There isn't a method of directly setting different surface densities that I can recall. As is, your internal volume density is set to 1.0 which will have a significant effect on your analysis. I suppose if you wanted to try and work around the problem, you could set the volume density to 0, and use negative volume geoms to cut the surface and then analyze each section in isolation. That would be a mess, however, and I don't really recommend it. I suppose it would depend on how different the materials are in those regions and how much that difference matters when you consider the weight of the fuselage structure, etc. If the mass or CG location is only changing by one inch, for example, the pilot shifting around could alter it more than that.