Frankly, your model doesn't illustrate any such thing. First, as has been noted, it's so grossly oversimplified as to be virtually worthless.
Second, it doesn't show any significant stress in any of the other floors, except at the floor "connections" of the one immediately below the top. And because you haven't actually modeled the connections, or used the proper materials, we can't draw any conclusions about that.
Third, as has been mentioned numerous times, and Minadin reiterated, the weak link in the entire structure is the floor-truss connections to the support columns. We know that these will fail first, but your model assumes that they won't fail, and grossly misrepresents the geometry of the contact area between the floors and the columns. Therefore the stresses shown in the other floors may or may not (most likely not) be accurate.
Finally, I'm not certain what you believe your model is supposed to prove. It seems as though you might think that if you've loaded the top floor with, say, 1,000,000 kgf, that not all of that force will affect the floor. But it will. Assuming the truss connections don't fail, there will be a a total of 1,000,000 kgf in the connections pushing against the support columns. However, by Newton's Third Law, there is also a total of 1,000,000 kgf in the support columns pushing back against the connections. Further, again assuming the connections don't fail, for the structure to be in static equilibrium, there will be 1,000,000 kgf at the bottom of the support columns pushing against the bedrock, and 1,000,000 kgf in the bedrock pushing back against the bottoms of the support columns. So the fact that additional stresses are shown in other parts of the model in no way proves that not all of the force of falling debris would affect the top floor.
My two-year degree from Indiana Vocation Technical College is in "mechanical design," and my job title is "mechanical designer."