Update. 
1.e4 g6 2.d4 Bg7 3. Nc3 d6 4.Be3 a6 5.h4 Nf6 6.h5
Has been plaid more in practice. Especially important is that in the TCEC most of the engine games ended with a white win.

1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.h4 h5 6.Nh3 Nf6 7.f3 Bxh3
I did a lot of analysis on the lines here in the last couple of months. As I mentioned before if white knows to put the queen on e2 the threats of e5 and g4 are very difficult for black to circle. Here are a couple of examples of the concrete difference the queen placement makes:
8.Rxh3 Nbd7 9.Qe2 c5 10.O-O-O Qa5 11.Kb1 b5 12.e5 dxe5 13.dxe5 Nxe5 14.Bxc5
or
11...Rc8 12.g4

Since white can mess up the flow of the thematic c5 Tiger plan black may try.
9...b5 10.O-O-O
Here the variations become a maze of options.
10... Nb6 11. e5 Nfd5 12. Nxd5 Nxd5 13. e6 
I would evaluate this as a problem for black.
10... Qb8 11. g4 b4 12. Nb1 
Where after extensive analysis whites' attack as well as a myriad of defensive resources such as a4,c4 and a rover from the h3 rook look to tip the scales in whites' favor.
10...b5  11.Na4 c6 12.Kb1 Nb6 13.Nxb6 Qxb6
Ends up really complex with white having ideas like g4 and d5 as well as ideas to mess with light square control on the queenside: a5 is sometimes met with the intrusive d5->Qb5. White has some waiting moves such as Rh1, g3 and Qd3.
The position is truly bizarre, in how both sides should probably make positional baby steps to avoid counters... 
I haven't found an equalizer yet, some of the positions are still a bit over my head. 


The closest to equality I think I've gotten in my analysis is:

7...b5
7.e5 Nh7 8.f4
White misplaces the knight and builds an Austrian center. Nd7 would result in e6->Ng5 and pain.
8...b4
And I think that the c5 central explosion responds adequately. Most challenging is propably:
9.Ne2 c5 10.Nf2 cxd4 11.Nxd4 dxe5 12.fxe5 
After which the position hangs on the edge of a knife.

7.Ng5 O-O
8.e5 Nh7
And a mess ensues yet again. The plans however remain. Black looks to destroy the center, and white tries to keep it together long enough to strike through on the black king. With optimal play, things look fine for black.

8.g4
Both c5 and Nc6 turn out interesting, only mentioning this here since it's somewhat forcing.
8.Qd2 c5
With white building up to e5->e6, g4 and O-O-O black has to strike. This pawn sac has some shades of the KID Sämisch c5 line. When plaid optimally black pieces simply outperform white relying on a large variety of ideas. Like the Sämisch this one requires some very concrete study since the comp is often less than obvious with stuff like Bxc3 and timing-sensitive knight maneuvers. There are a couple of high-level draws.

I've been studying this line for some time now too.
I'd agree that avoiding h4 h5 is best for white in these positions but there are some cool counter-attacks against Bh6, and g4 and Nf4 have their respective counters in h5 and e5.


I think another problematic move order for the b4 plan is 
1.e4 g6 2.d4 Bg7 3. Nc3 d6 4.Be3 a6 5.h4
The point is if 5...Nf6 6.h5 is powerful sacrifice because the queen is still on h5 6.Nxh5 Rxh5 7.gxh5 Qxh5 with black having a hard time with direct threats of Bc4 and Nf3 -> g5 and the concrete problems with trying to arrange the pieces for a queenside castle. I've vetted the line with leela and stockfish to the best of my abilities and it seems pretty grim.

Now of course black could play
5...h5 6.Nh3
In the case of 6...Bxh3 (or 6...Nf6 7. f3 Bxh3) white seems to have a pretty dangerous idea in Qe2. The move does a good job pointing out blacks lack of light-square control and placing the rook soon to be on d1 opposed to the black queen. Due to these factors alot of the black lines are spoiled by the e5 and g4 pushes or white's lacking control over the queenside... Black probably has something playable, but I've not been able to find it yet.

Now, if black doesn't take on h3.
6...Nf6 
(if 7.Ng5 Ng4, if 7.Qd2 simplest is 7...b5 where 7.f3 Bxh3 is line known to work out ok for black, and 8.Ng5 Ng4 9.Bf4 Nc6 10. f3 e5 gives a complicated game with pretty good prospects for black)
7.f3 b5 8.Ng5    
and white's options to go kingside or stay in the center make b4 a lot more double-edged. Still 8...b4 9. Nce2 is interesting with some lines like 9... O-O 10.Bc4 and 10.g4. 

Another line I'm still working on figuring out is 7... Nbd7

Hi.

There are seven correspondence games with 8...b4 (twenty for 8...c6, eighteen for 8...Nbd7). 9.Nce2 is the move tried in response. In principle white wants to get his g1 knight to g5 and the e2 knight to f4. One of the questions the correspondence players seem to have debated is if (or more likely when) to play Kb1. Some do it early. A provisional mainline would seem to be the following sequence up to 16.Nd5. White moves pieces to where they want to be and starts to put pressure on the opponent somewhat. At that point black has a big choice, but it looks to me like the position holds more danger than might be thought and he should just go with the move seen in practice: 16...c5. There are a fair number of alternatives earlier on. For example I am not exactly sure what white does if black just takes the knight on h3 as soon as it appears.

1.d4 g6 2.e4 Bg7 3.Nc3 d6 4.Be3 a6 5.h4 h5 6.Qd2 Nf6 7.f3 b5 8.O-O-O b4 9.Nce2 a5 10.Nh3 Nc6 11.Ng5 Ba6 12.Nf4 Bxf1 13.Rhxf1 e5 14.dxe5 Nxe5 15.Bd4 O-O 16.Nd5 c5
16...Nxd5?! 17.exd5 Bh6 18.g4 hxg4
18...f6 19.f4 fxg5 20.fxe5 dxe5 21.Bxe5 Rxf1 22.Rxf1 Qe7 23.Re1 Re8 24.gxh5 gxh5 25.Bg3 Qf8 26.hxg5 Rxe1+ 27.Bxe1 Bg7
19.f4 Nf3 20.Nxf3
20.Rxf3 gxf3 21.h5 Bxg5 22.fxg5 Qe7 23.Re1 f2! 24.Rxe7 f1=Q+ 25.Re1 Qf3
20...gxf3 21.Rh1 Qe8
21...b3 22.cxb3 a4 23.h5 axb3 24.hxg6! bxa2 25.Kc2 c5 26.dxc6 Qc8 27.Rxh6 Qxc6+ 28.Qc3 Qe4+ 29.Qd3 Rfc8+ 30.Kd2 Qxf4+ 31.Qe3 Qxe3+ 32.Kxe3 Re8+ 33.Kd3 fxg6 34.Rxg6+ Kf8 35.Ra1
21...a4 22.h5 b3 23.Kb1 Re8 24.hxg6! Re2 25.Rxh6 f6 26.Qc3 bxc2+ 27.Kc1 cxd1=Q+ 28.Kxd1 Qf8 29.Rh1 Qg7 30.Qxf3 Rae8 31.Qg4 a3 32.Rh7 Re1+ 33.Kc2 R1e2+ 34.Kc3 axb2 35.Rxg7+ Kxg7 36.Qh4 b1=N+ 37.Kb4 Rb8+ 38.Kc4 Na3+ 39.Kd3
22.Rde1 Qa4 23.h5 b3 24.cxb3 Qb4 25.hxg6 Qxd2+ 26.Kxd2 Bxf4+ 27.Kd3 Be5 (+=) White has a slightly better endgame

17.Bxe5 dxe5 
18.Kb1 c4 19.g4 Nxd5 20.exd5 c3 21.Qd3 hxg4 22.fxg4 Qd6 23.h5 a4 24.Ne4 Qa6 25.hxg6 Qxd3 26.cxd3 cxb2 27.Kxb2 Ra5 28.gxf7+ Rxf7 29.Rxf7 Kxf7 30.Rf1+ Kg6 31.d6 Rd5 32.Kc2 Rd4 33.g5 Rd5 34.Re1 a3 35.Rc1 Bf8 36.Rf1 Bg7 37.Rg1 Bf8 38.Rh1 Rb5 39.Rf1 Bg7 40.Kb3 Rd5 41.Kc4 Rd4+ 42.Kc5 b3 43.axb3 Rxd3 44.Kc6 a2 45.Ra1 Bf8 46.Rxa2 Bxd6 47.Nxd6 Rxb3 48.Rg2 Rc3+ 49.Kd5 Ra3 50.Ke6 Ra5 51.Nc4 Ra6+ 52.Kxe5 Ra4 53.Kd5 Ra8 54.Rg1 Re8 55.Ne5+ Kf5 56.Nf7 Kg6 57.Rf1 Re2 58.Rf6+ Kh5 59.Kd6 Rd2+ 60.Ke7 Re2+ 61.Kf8 Rg2 62.Kg8 Rg1 63.Kh7 1-0
Soares, Francisco Antonio de Freitas  (2100)   --   Bobone Carvalho, Diogo  (2006)
POR/CNA/pr14  3 (POR) (-)  ICCF
2020.12.20  1-0

18.g4 c4 19.Kb1 Nxd5 20.exd5 e4 21.Nxe4 b3 22.cxb3
22.gxh5 bxa2+ 23.Ka1 c3 24.Nxc3 a4! Should give enough counterplay
22...cxb3 23.a3 Rc8 24.Nc3 Bxc3 25.bxc3 Qd6 26.gxh5 Qxa3 27.hxg6 fxg6 28.Rg1 Kh7 29.h5 b2 30.Rxg6 Qa1+ 31.Kc2 Qa4+ 32.Kb1 Qa1+ 33.Kc2 Qa4+ 34.Kb1 Qa1+ 35.Kc2 Qa4+ 36.Kb1 Qa1+ 1/2-1/2

Franzén, Thomas  (2340)   --   Pospísil, Ludvík  (2479)
VWC12 /pr 2 (-)  ICCF
2019.09.01  1/2-1/2

Hello.

MartinC wrote on 08/21/22 at 06:58:21:

Slightly lost perhaps. I would think most times black will have played Bb7 or Nd7 instead of Nf6. If Bb7 is in the position instead of Nf6, then this whole b5-b4 idea becomes a bit weak. The bishop basically is just misplaced. If Nd7 is in the position instead of Nf6 then black can still play this b5-b4 idea, it's just not so impactful.

1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.Qd2 b5 6.h4 Nf6 
Is very much a line though. Not to mention 1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.h4 Nf6; even though I pointed out some potential problems with moving the knight in this instance a while ago. In the position after the bolded text... If black goes directly for 6...h5 white can suck a bit of life out of the position with 7.a4 b4 8.Nd5 a5 9.Nh3 Bxh3 10.Rxh3. It's probably not so easy to get play there for black. Also Tiger asfair mentions 7.Nf3!? as well.
After the knight move black keeps the flexibility from not having played Nd7 or Bb7 at least a while longer. 
7.f3 h5 8.0-0-0
Is a reasonably common continuation and Tiger likes 8...c6, as mentioned. The basic issue is if black plays the generally desirable 8...Nd7 (or even 8..Bb7) is that both make capturing on h3 if white goes Nh3 very hard. A new move for black is in principle quite welcome for this reason

It's notable btw that Fernandez recommends
1.e4 g6 2.d4 Bg7 3.Nc3 d6 4.Be3 a6 5.Qd2 b5 6.0-0-0 b4
and Jones as far as I can see misses this in his white repertoire. Maybe he was made aware and liked the move there. Such can, plausibly at least, be a reason for him going 6.h4 this time.

Have a nice day.

Hi.

Looking at chesspub.com August 2022 update you see Pirc/Modern expert Gawain Jones, with white, face a certain M. Carlsen. I recommend taking some time to appreciate the game:
https://www.chessgames.com/perl/chessgame?gid=2281509

Carlsen plays the active 8...b4, somewhat circumventing the very established 8...c6 and gets an interesting position. The plan against the likely more critical 9.Nce2 would presumably be to put a knight(!) on c6. The 8...b4 move is rare, yet when it was played it seems to have been played mostly by strong players. I'll check a corrbase later and see if it's seen action there as well. I checked some Pirc and Modern books earlier and can't really see 8...b4 mentioned; although Fernandez mentions an early b4 move in a similar position. 

White can consider different move orders to try and get around the position after 8...b4 in the Carlsen game. The same Jones for example goes long castling before h4 in his coffeehouse repertoire. Doing this would likely mean black will not want to play h5 and then he needs to find some other move and even if he goes b4 the move after it's not going to be the same position.

Have a nice day.

Edit: spelling.