Well, Pakistan has nukes, so strictly speaking it is a more important country to keep stable. However, it isn't clear that it can't take care of itself. While pressure in Afghanistan could result in less deaths in Pakistan, I don't think it is needed to prevent Pakistan's government from collapsing. As regards Afghanistan, continuing to suppress Al Qaeda's operational capacities is of course a goal. But I would argue that in the long-run turning Afghanistan into a modernizing instead of a stagnant country would, if it were possible, be the best way for the US to save face. In humanitarian terms, owing to lowered starvation rates, this would more than vindicate a conflict that has thus far been far less destructive than the one in Iraq (thus far in Afghanistan civilian casualties are between 12,000 and 32,000, Iraq was 500,000 to 1.4 million). The increased emphasis on preventing civilian casualties makes it likely collateral damage will decrease even more, at least from our side. Al Qaeda is a threat to be sure, but it is often over stated. Al Qaeda is only very dangerous if it procures WMDs. 9/11 killed over 3,000 people, but 10,000 people in the US die a year of homicide, and over 15,000 people die in the world every day of starvation. The subsequent attacks in Europe were even less deadly. It would be more cost effective for us to simply focus on domestic security and some bombing. Staying in Afghanistan is therefore against our own self-interest, but I still think we have an obligation to facilitate the country's transition. If we can't succeed therein then we should leave now. But a lot of people thought the effort in Iraq was doomed too. We pushed a bit harder, and so achieved our goals and thereby relieved ourselves of the burden. The same might be possible in Afghanistan, and if a surge didn't work than we could leave knowing we had exhausted our options.