In my view 5...c6 would be ideal against this system, but not if White continues to play in 150 Attack style (i.e. does without a4 and plays Bd3 in response to ...b5)
In answer to your questions:
(1) It may well be that 6...b6 7.e5 is strong. Since (after 7...Nfd7 8.e6) White is likely to use his h-pawn as a battering ram in any case, he has effectively saved a tempo by not playing h2-h3. I suppose you could argue that the tempo Black has lost is castling, and that the king is safer on e8 than on g8. I doubt this, though. It is not as if it is going to escape to the queenside any time soon!
(2) It doesn't really make any difference as far as White is concerned: 1.Nf3 d6 2.d4 g6 3.e4 Bg7 4.Nc3 and if 4...c6 or 4...a6, he simply plays 5.a4 when I don't see how Black can do better than 5...Nf6.
To my mind, the fact that White has committed to Nf3 is a good reason to play a Pirc move order rather than a Modern, as it rules ut his two most dangerous options against the Pirc, namely 4.Bg5!? and the Austrian Attack.
(3) You know very well my feelings about the Dutch Defence!
(4) Davies does indeed recommend 1.Nf3 g6 2.e4. Besides, 1...g6 does not fit in with my repertoire after 2.c4
There are of course various ways to avoid all the nastiness in my initial post - I have already indicated that a favourable transposition to the Czech System is possible at move three, and Black can also play 1.Nf3 d6 2.d4 Nf6 3.Nc3 g6 4.e4 Bg7 5.Be3 a6 6.a4 0-0 7.h3 Nc6!? which does not lead to quite such sharp play.
But the two variations I outlined above have I think been Black's main responses to this Be3/h3 system so I wanted to highlight potential pitfalls to fellow Pirc players. The variation has fallen out of favour with White players and has been seen as fairly safe for Black, so it is as well to know if White players are going to be coming at it from a new (and dangerous) angle.