No. If you actually read the paper linked (and the Supplementary Online Materials, which contains a lot of the actual information), it is shown that rake angle is not necessary for the bike to be self correcting. You can build a bike with a negative or zero rake angle that is still self-stable (at least, according to the bicycle dynamics modelling software they were using, JBike6; they didn't actually build this particular bike, but did build one that had small negative trail and no gyroscopic effects that was still stable).
As they demonstrate in the paper, none of rake angle, trail, or gyroscopic forces are either necessary or sufficient for self-stability. All of them can influence stability, so saying the gyroscopic theory has been proven wrong is not entirely right either; it is a part of the dynamics that adds stability, it is simply not necessary or sufficient on its own. In fact, the paper shows that on the "benchmark bicycle", removing the gyroscopic force makes it unstable, so on that particular design, the gyroscopic force is necessary for its stability (thus explaining why it was believed for so long that gyroscopic force is what provided stability).
What we know is that gyroscopic forces, trail, rake angle, and distribution of the center of mass of the fork and body of the bicycle all influence stability (in particular, the center of mass of the body of the bike being substantially higher than that of the fork); none of them alone are sufficient to provide stability, and likewise none of them alone are necessary as we can build bikes without them that are still self-stable.
> > Having a rake angle on the front wheel makes a bicycle self correcting ..
> No. If you actually read the paper linked ...
Please don't start replies with a "no", especially when you don't disagree! You reply that it "is not necessary" which does not negate your interlocutor's point!(probably going to regret going meta, but the initial 'no' in forums and irc really bothers me)
Also, while a riderless bike is fairly stable, a rider overpowers the autocrorrective nature when they hold the handlebars. So, really bikes are stable in large part because the rider balances for the bike as seen by your ability to keep a bike upright without moving and it becomes far easer to do so at as you speed up.
The takeaway is that there are several factors which influence the stability of a bike. We know of certain designs which utilize one or more of these factors to achieve self-stability (and the conventional bike has all of these factors, hence why it tends to work so well), but we don't know the exact set of conditions on the combination of factors which would allow you to characterize which designs are stable versus unstable, without simply trying out any given design and simulating it.
They very clearly describe (at the end of the video) that even with gyroscopic and tracking forces removed, the two most important factors are where the center of gravity is, and the tendency to turn into a fall (arguably just the second, but it is caused by the location of the center of gravity with regards to the pivot).
[1] http://www.youtube.com/watch?v=YdtE3aIUhbU
Edit: more info on the TMS (two-mass skate) bike at its wikipedia page: http://en.wikipedia.org/wiki/Two-mass-skate_bicycle
A bicycle in motion adjusts its center of gravity to remain upright. It's very similar to the inverted pendulum problem.
Look at a bicycle directly from behind with the wheels exactly lined up. Now imagine that you could frictionlessly slide the two tire patches left and right. The similarities to the inverted pendulum become more clear.
Of course, it is more complicated than the classic inverted pendulum. Instead of one point of contact under the mass, there are two. And the two points of contact (i.e. the wheels) have their own complex dynamics.
Having a rake angle on the front wheel makes a bicycle self correcting (if the c.g. is on the right side of where the wheels contact the ground, then a right turn is induced in the front wheel by the rank angle)
There are two major forces that must be in balance to turn a bicycle - the side force from being off center with respect to c.g., and the centripital force in the turn.
Ever watch a cyclist train on rollers? That's much closer to an inverted pendulum. And since there is no forward momentum, there is no centripital force, which makes it more difficult to remain upright on rollers than on pavement.