The bigger mess 'theory' is a rather simple rule of thumb, I'll grant. It's attractive because it can be tested, to a degree.

What you want to know, of course, is what course of action would have been best, which would require plotting various scenarios, which is a lot of work and largely untestable. But the bigger mess rule ignores the hypotheticals that a situation is on the brink of spiralling beyond control and, on the opposite, that a better solution is clearly available.

All ethnic Serbs are not yet out of Kosovo, though the majority fled.

Though an aerial bombardment has an effect of rallying the people around the leader, losing a war does not have the same effect. Kosovo was the heart of Milosevic's nationalist politics. Posing as the protector of the Serbian minority there is what brought him to power, and failing at this will have had an impact on the elections. The credit for deposing Milosevic goes to the Serb people, but the Kosovo war will have played a role in it.

The Serbian economy recovered to its pre-war level in 2002 or 2003 and has since grown by 7% in 2004 and 5.9% in 2005.

The lawlessness and banditism in Kosovo was already there before the war began. In addition there was a guerilla war, which was threatening to spill over into Macedonia.

In total, I don't think it is arguable that the overall situation now is not better now than before the NATO intervention. What parts of the overall improvement were caused by the intervention can be argued about, as well as whether the intervention was worth it and whether the situation could not still deteriorate.