If anyone did those jumps (changing the quad Salchows to quad toes), he would be Plushenko, and it wouldn't matter what anyone else did or how uneventful his prigram was otherwise.
Many thanks to gkelly for the clear breakdown of the relative programs and analysis.
To address Mathman's point, among the current men, that pretty much means Plushenko only. Of the other men:
1. Those with consistent quads: Joubert, Lambiel, Sandhu, and Klimkin are inconsistent with their 3A's, and don't perform 4/3 combo, 4 solo, 3A/3 combo, 3A solo. The different between a well done quad combo and a messy 3A (without fall):
4T/3T: 13.5 +1 GOE: 14.0
3A, -1 GOE: 6.5
Total 20.5
3F/3T: 9.5 +1 GOE: 10.5
3A, +1 GOE: 8.5
Total: 19
or
3Lz/3T: 10 +1 GOE: 11
3A, +1 GOE: 8.5
Total: 19.5
Li and Sandhu are as likely to perform a 4T/2T as a 4T/3T, which would mean:
4T/2T: +1 GOE: 10.8
3A, 0 GOE: 7.5
Total 18.3
The difference between a second 4T and a second 3A is 1.5 points.
2. Those who've been without quads or consistent quads: Buttle, who has a dodgy 3A, and I can't remember when he performed two flawless 3A's (one in combo) in one LP. (Please tell me if he has; I'd love to see it.) His consistency with the Lutz has been questionable. Weir doesn't use all of his jump passes, but he's had problems on the solo 3F in a number of LP's. It's hard to say whether he could sustain the overall quality of his jumps if he did the maximum, because we haven't seen him jump late very often. Oda had late-program issues, probably stemming from a combination of nerves and the transition from junior to senior length programs, and Takahashi can be all over the map. Lysacek has issues with Axels.
If I've seen Preaubert in person, I don't remember, and if he had great skating skills, I would have noticed. On the small downloads, he seems solid and a bit stolid to me, and the quality of his jumps is mixed.
That the quad doesn't get more points, lifting the value of a quad in combo much higher than a 3F/3T, 3Lu/3T, even when the combo is a 4T/2T vs. a 3A/3T, is a much-argued point.