Hello again I might have misunderstood something then.
When I used salign with several templates I noticed that the *fit_fit* files have new coordinates. I assumed that MODELLER tried to place them in the same coordinate system according to the sequence alignment produced by salign to get "better" starting positions before energy minimization.
I have compared the different *fit* "clusters" and, to me, it seems like the first fit sort of places the main (top) template in some sort of origo and the following fits (*fit_fit*, *fit_fit_fit*) align the other templates with the main one according to the sequence alignment produced by salign.
Due to these observations I assumed that you had to transpose the template pdb files so that they were sort of placed "correctly" with respect to eachother.
The first FAQ entry (as you probably know :) ) says:
"If no additional information is available about the relative orientation of the two domains the resulting model will probably have an incorrect relative orientation of the two domains when the overlap between A and B is non-existing or short. To obtain satisfactory relative orientation of modeled domains in such cases, orient the two template structures appropriately before the modeling."
But with sufficient overlaps you think I can just use the original template pdb files and disregard the fit steps performed by salign?
Thank you very much Niklas