So what does Cohen have to say about the 4.Bc4 variation of the Four Knights?
In the line 4.Bc4 Nxe4 5.Bxf7+ Kxf7 6.Nxe4 d5 7.Ng3 e4 8.Ng1 Cohen recommends 8...h5. The computer will tell you that this is better than the 8...Bc5 move I suggested. The best line for white given by Cohen is 9.d4 exd3 10.Qxd3 Nb4 11. Qd1 Qe7+ 12.N1e2 h4 13.Nf1 d4 14 a3 h3 15.gxh3 Nc6 (fig 1)
Fig 1. Position after 15...Nc6
An important line covered by Cohen is 4.Bc4 Nxe4 5.Nxe4 d5 6.Bb5 dxe4 7.Nxe5 Qg5 8.d4. (I had discussed only the 8.Bxc6+ and the 8.Nxc6 lines in my second posting on this opening.) 8.d4 is undoubtedly objectively best, but in my experience rarely played. Here play continues 8...Qxg2 9.Rf1 and then 9...a6!? (fig 2) in an effort to force the bishop off of the a6-f1 diagonal so that black can win the exchange after playing ...Bh3. I will not give the rest of Cohen's analysis, which is lengthy and nearly without verbal comment. (In any case, remembering the 9...a6 idea is probably all you need here.) Black is substantially better in all lines.
Fig 2. Position after 9...a6.
Cohen also analyzes the 4.Bc4 Nxe4 5.0-0 line, which I did not cover. Here he makes the case for black forcing white to prove its point. After 5...Nxc3 6.dxc3 black should play 6...h3!? so that he can safely reply to 7.Qd5 with 7...Qf6. (Curiously, my engine recommends 6...f3!?, a true computer move, weakening the light squares around the king.) Cohen argues that black is at least equal in all lines. Best for white after 6...h3 is 7.Re1 d6 8.Nd4, but black can come out slightly better if he plays 8...Na5 (fig 3), intending ...c5 after the bishop moves.
Fig 3. Position after 8...Na5.
One notable absence from Cohen's book is any mention of Larry Kaufman's inventive answer to the 4.Bc4 Nxe4 5.Nxe4 d5 6.Bd3 line: 6...Nb4!? It shows that Cohen's work is somewhat short of definitive on the 4.Bc4 variation of the Four Knights.
In conclusion, I think Cohen's book is a bit better than Hansen would have it. The seemingly useless analysis of lines like the 4.Bc4 variation of the Four Knights is helpful for "class" players (below 2000 USCF). And, although Cohen surrenders to the temptation to give long computer lines, some of his explanation and analysis of important Petroff variations is excellent. His discussion of the ideas in the 3.Nxe5 d5 4.Nf3 Nxe4 5.d4 d5 6.Bd3 Nc6 7.0-0 Be7 8.Re1 Bg4 9.c3 variation is, for example, the best I have ever seen, and his analysis of it seems to offer new prospects for the pawn sacrifice and rook lift line that begins with 9...f5 10.Nbd2 0-0 11.Qb3 Kh8 12.Qxb7 Rf6. Petroff players will want to study this book. Although Sakaev's Petroff: An Expert Repertoire for Black is probably better by most standards, Cohen's book is an useful addition to the literature on this opening.