Two more slams? Which two slams do you people think he will win? Sure, Sampras bagged a late one, but he had big weapons that were independent from one another. If the movement wasn't great, the serve and forehand were still enough to get the job done. You can't say the same thing about Federer. I'm not sure he'll win two more. I'm not writing him off yet, but I think his chances are much better at the USO than anywhere else. I'm still baffled that he was able to do what he did to Murray there last year (that was an upset IMO). I just don't see him winning Wimbledon again. History is against him. Typically when multiple-Wimbledon champs lose, they do not win again. And, I just do not see him beating Nadal 2.0 on grass.
Sampras won Wimbledon 93-95, then lost. Then he won 97-00. That directly negates what you say about the history of Wimbledon champs and such. On top of that, he has lost a total number of sets per year:
2006: 1 (1 to Nadal)
2007: 3 (2 to Nadal)
2008: 3 (3 to Nadal)
So he has lost 11 sets to (6 to Nadal) in the last 41 matches (133? sets). Sure, Murray is a new potential threat and Fed is no longer '05, but its wishful thinking to close the Wimbledon books on him right now. Should Nadal get knocked out, he immediately becomes the favorite by a decent margin. If he finds any consistency in the next two months, he may still be the favorite against even Nadal.
Wimbledon is probably his best shot for the coming years, especially if the Murray factor somehow convinces them to speed the grass up a bit. USO as his second option makes me fairly confident about his chances to get two more. Making four Finals and one Semi playing as he has since AO '08 is a further confidence boost.
I don't think Fed gets #1 back. To do that, he has to win masters series events, and given the recent additions to his life coupled with his more frequent health struggles and the emergence of even more young talent, I don't see him making a big push to keep winning the smaller stuff when he's two slams away and the clock is ticking.