Thanks for your response.  So me along with my engine were correct about Korchnoi allowing any advantage (largely due to the extra pawn in the engine's evaluations I'm sure) he may have had slip away particularly with 20. Qc2.  So White will look for improvements, therefore it is prudent for Black to do likewise -- whether or not 13...Rg8 is or isn't is yet to be determined.

The following line is from last years game Korchnoi-Vallejo Pons:

1. d4 Nf6 2. c4 e6 3. Nf3 c5 4. d5 b5 5. dxe6 fxe6 6. cxb5 d5 7. Bg5 Qa5+ 8.
Qd2 Qb6 9. Bxf6 gxf6 10. e3 Nd7 11. Nc3 f5 12. a4 Nf6 13. Be2 Bd6 14. O-O Bd7
15. Rfd1 Ke7 16. a5 Qc7 17. h3 Rhg8 18. Nh4 Rab8 19. Kh1 Be5 20. Qc2 d4

Black wins in another 15 moves but according to my engine is only about equal here.  But before Korchnoi's last four moves starting at the 17th, my engine considers the position more or less +/= for White.

Going back to move 13 however 13...Rg8, putting pressure on g2 before White castles short, may be an improvement that comes close to equalizing.  At least according to my engine.  I'd be interested in any annotations that anybody can provide from sources they may have that I don't, such as Informant.