It matters more for researchers, yes, but not as much as you may think.
The answer is yes, it tends to work out that way because as a whole the individual researchers at Wisc-Mad (and similar schools) tends to be stronger. THIS is what schools will look at.
So if Wisc-Milwaukee has a fantastic researcher who is publishing tons of GOOD work (that's important to keep in mind since some people have tons of publications but when you look at it, it isn't very good work, and most of the publications are in unknown journals), than in most cases, you are better off going there than with someone who isn't doing as much at Wisc-Mad (though knowing the school, I don't think they HAVE any faculty that aren't doing much!).
So basically, people care where you went to school, but not as much as what lab you came out of. Because really, what does the "school" do for you? Classes? Ask any grad student, classes are relatively minor, relatively unimportant parts of graduate school.
If all other things are equal, than yes, I would choose Wisc-Mad over Wisc-Mil. The point I think we're all trying to make is that unlike law or mba programs, ranking should not be a major factor.
My personal recommendation (its what I did) is to look at faculty ONLY first. I used rankings as one way to examine the likely productivity of my faculty of interest, but that was about it. You can narrow it down later, but I think people (at least on the research side of things) ought to reframe the grad school search as a "lab search" with the grad school being secondary.