Warning: I may or may not be full of **** in the following post.
I *think* you're gonna want a lot more than a 1mm thick base to a block. Unless the majority of the cooling is happening right above the core, the heat is gonna have to conduct through the copper base to the rest of the waterblock. In my case, I'm working on a #rotor style multi-hole block. I whipped up a quick 3d model of the block, with 3/16" holes spaced in a quarter and 1/16" channels between the holes. I then took it to a school computer/design lab, and ran a basic FEA optimization for the thickness of the base of the block (Have I mentioned that my high school rocks?). I applied a 75 watt load to a CPU-core sized rectangle on the base of the block, and a simple convection boundary to the interior of the block (there was no way to simulate flow throught the block, but I think that may be irrelevent for what I was trying to find out). The optimum thickness for the lowest temperature on the core was just over 3/16".
None of the above applies, of course, if the block you're making cools mostly through squirting a jet of water right above the core. In that case, make sure there is as little as possible copper between the water and the CPU.
Anybody that really knows what they're doing: Please go ahead and correct me wherever I may be wrong. Or affirm what I just said... that would be better.