Honestly, if the jump from IE6 to IE7 is too much for some users, the jump from IE6 to IE8 going to be even more frightening. Remember IE7 didn't really fix any IE6 rendering issues, still some (badly designed) websites had issues with the transition. Now that IE8 compliant by default, there's going to be a lot more web content that looks and works slightly differently.
Basically the IE6 user base is upgrade averse as it gets, and a new major revision never cures that. IE6 can only be eradicated by a coordinated desupport effort, and that's a total prisoner's dilemma mess. No one's going to risk their traffic for it.
It seems to me that IE6 is capable of everything that newer browsers are (including new IE's). The problem is that in some cases IE6 requires more or different effort from the content creator to get the same effect as other browsers. I mean, what is something that is just plain impossible to do in IE6 that we will be able to do in Firefox 3.5?
If IE6 has the UI that people accept, and if IE6 has the performance/resource utilization that those people like (compared to modern browers, it is pretty lightweight except on javascript-laden pages), and if IE6 already works for every site that they use, why should they switch? Why should they buy a new computer or more RAM just to run a new browser when their current one works just fine? Why should they download a third-party browser and then have to deal with even more updating and migrating their bookmarks and learning a new UI?
We need to understand that there are going to be a significant number of people running IE6 on Windows XP even a couple of years from now. Nobody has created a compelling reason for them to upgrade and it doesn't seem like it will happen until Microsoft stops issueing security upgrades for those products. Even then, a lot of people won't care if they get security upgrades or not. (I know of a small business that is using a ten-year-old desktop running Windows 98 as their only file server today.)
Basically the IE6 user base is upgrade averse as it gets, and a new major revision never cures that. IE6 can only be eradicated by a coordinated desupport effort, and that's a total prisoner's dilemma mess. No one's going to risk their traffic for it.