Glenn, I did reply to a question about this in the 8...Na5 thread.
Due to the recurring ignorant putting down of things I have written elsewhere I had decided not to contribute more to this forum. But now I think I will just ignore this kind of insults from small people with huge ego problems.

Yes, there was a series of two articles by Jensen in NIC Yearbook 68 and 70 on the line 8...Ne5 9.Nxe5 dxe5 10.e4 f4 11.b4 and only on this line. I did not say anything about 11.gxf4, 10.Qb3 or any 9th move alternatives.

Personally I have a good score against 10.Qb3 but only because White players played the wrong moves.
The problem with 10.Qb3 h6 or 10.Qb3 Kh8 is that White can reply 11.e4! and if 11...f4 12.gxf4. In these lines Qb3 is clearly more useful than Kh8 or h6. A Black Nxc3 does not hit the queen on d1 anymore (compared with Chandler-Plaskett) and Qc3 check becomes a possibility after Kh8.
Ehlvest even gives this critical line (on page 67) without realizing how imprtant it is.
The problems were convincingly highlighted by Robin Edlund in a letter to the Forum section of NIC Yearbook 71. I checked all of his analysis and found no holes in it, and neither did Sosonko. Unfortunately the editorial team did not forward Mr. Edlund's letter so I was denied the chance to reply.
Given that the move 11.e4 is the problem, Black has only two improvements:
A) 10...e4 is the obvious choice. This is not normally a move one wants to make if White can reply 11.f3 but there are extenuating circumstances for Black.
B) Perhaps Black can use Kh8/h6 in one of the ...e6 lines (as in 10.e4 e6 which is better for White)
10.e4 e6 11.Qb3 Kh8 is given as better for White by Elvest (Donner-Johannesen, Beverwijk 1965) and 10.Qb3 Kh8 11.e4 e6 reaches the same position. However Luccioni improves on that game with 12.Be3 exd5 13.cxd5 f4 transposing to Uhlmann-Espig, DDR 1976 or Tartakower-Alexander, Hastings 1953/4. He also gives the possible improvment for White: 14.gxf4 This is a position where White has the resource 14...exf4 15.Bxf4 Nxe4 16.Bxc7! since Black cannot reply 16...Nxc3.
10.e4 e6 11.Qb3 h6 looks worse for Black, though it does prevent Bg5 in reply to ...e6.

In conclusion, I would play 10...e4 and also take a good look at 10...Kh8 11.e4 e6!?

Strangely, these posibilities are not examines in YB 71.

I am quite interested in this line, and I would play it if I thought it were sound.  There was a two-part article in recent NIC Yearbooks trying to rehabilitate it.  The main lines involve banzai attacks against the White king.  Very risky, but one way to play for the full point, and entertaining. 

But in YB 72 (the most recent) somebody wrote in to say that 9. Nxe5  dxe5  10. Qb3 was much better for White, presenting some backup analysis.  Sosonko agreed.   

I haven't had the chance to look at it myself.

Do I remember the recent New in Chess yearbooks running a few articles on this line? You might like to check. 

This seems like a fun line for black, although it looks ugly.