Does the observational process quantum-->classical ever reverse?
That's a very controversial question because it depends on exactly what you mean by quantum-->classical.
If you take the view that quantum mechanics applies to everything then there really is no quantum-->classical transition and there is no fundamental reason why the transition that occurs in a measurement could not be reversed. However, there is a practical problem in that you would have to be able to control the quantum mechanical degrees of freedom of very complicated objects, including the measuring device, the environment of the measuring device and possibly the observer as well.
However, if you take the view that wavefunction collapse represents some real physical process, then it would not be possible to reverse the transition that occurs in a measurement even in principle.
If you take a Copenhagenish view, then a measurement is simply a 'thing' that casues an irreversible quantum-->classical transition and hence it cannot be reversed by definition. If you found something that you say is a reversible quantum-->classical transition then a copenhagenist would simply deny that such a thing could be regarded as a measurement in the first place. It is a common copenhagen strategy is to frame things in a way such that seemingly interesting questions actually have no meaning. Depending on your view, this is either an extremely useful way to think about quantum mechanics or simply plain nonsense (I opt for the latter).
At present there is no clear-cut experimental way to resolve this issue and clearly no way to rule out the Copenhagen view in any case. However, as the technology for coherent control of larger and larger systems is developed then it should be possible to at least make one of the other positions seem incredibly unlikely.
Even in the Copenhagen interpretation, it seems possible to "reverse" wave function collapse in a sense. An isolated system once measured can't be "unmeasured," but nothing is isolated. Let it interact with something, and you'll have a mixed state again. Now separate everything out again, and you can repeat your experiment. The environment will be slightly changed though...
From what I remember, "environment induced decoherence" would prevent the "undoing" of a measurement. However, I have seen articles about a "quantum eraser". I think one of the authors of this article was a guy named Scully. You can do a search on this. I think there may be other articles that include references to a "quantum eraser" or "quantum erasure".
"Environment induced decoherence" prevents the undoing of a measurement in a practical sense, as I was trying to explain in my previous post without introducing the jargon. Actually, it is the same sense in which one cannot reverse a transition to thermal equilibrium. For example, suppose you have a box with heat-proof partition in the middle with cold gas on one side and hot gas on the other. When you remove the partition, the gasses will mix together and you will end up with a box full of warm gas. It is theoretically possible to reverse this transition, if you could control all the microscopic degrees of freedom of the gas particles, but in practice it is almost impossible.
This is true in an interpretation that says everything can be described by quantum physics, which is assumed by most modern physicists. On the other hand, Copenhagen assumes that the measurement process cannot be described by the unitary dynamics of quantum physics and so a copenhagenist would simply deny that what you have described constitutes a measurement. Simply put, if you have control of all the quantum degrees of freedom of a system, then that system is not a measuring device according to Copenhagen.
Separate names with a comma.