I'm reading Sherwood and Chabay's brilliant textbook Matter and Interactions, in particular the section that deals with how the surface charges in an electric circuit distribute themselves to generate the electric field within the wire.
It is possible that the surface charges are pinned at sites on the surface, but it is also possible that they are mobile. Even if they are mobile, their contribution to the current is infinitesimal, because I = JA = sigmaEA, and the cross-sectional area associated with the surface charges is completely negligible compared to the rest of the cross-sectional area. So whether the surface charges are mobile or pinned is irrelevant.
The surface charge distributions were calculated by a charge-field relaxation method described in the "Articles and talks" section of matterandinteractions.org. The VPython programs let you view these distributions and interactively explore the net field everywhere.
The number of electrons on the surface is EXTREMELY small compared to the number of electrons in the wire. Think of it this way: In a metal there is a "sea" of electrons that are not bound to atoms but are free to roam. If this sea expands by an infinitesimal amount, there will be a small (but essential) number of electrons on the surface, and the interior of the wire will remain very nearly neutral. The electrons that move to the surface do NOT travel a long distance to get there; there were already LOTS of free electrons right next to the surface. Sorry, but I don't understand your second set of questions. I strongly recommend that you study the paper whose link is given above.
c80f0f1006