You guys want some real info on this? I can contact some of the professionals I know regarding heat x-fer of gasses and liquids. Their educations range from England to USA to Australia and even Japan. Or I can sum up their knowledge.
Fact: increasing flow increases heat transfer. Period.Why? First, faster moving liquids and gasses have a smaller/thinner boundry layer. The boundry layer is the thin area between the surface and the liquid/air that doesn't want to move, and therefor not transfer heat. Reduce the boundy layer, and heat transfer raises by near-expotential amounts. The irregular coolant passages also lend to adding turbulance, which can reduce the boundy layer by a lot. If you want to experiment, wet your fingers and quicly put out a burning match with them. You wont' get burned [not badly, anyways
). The water emulates a boundry layer; it reduces heat transfer, and even insulates when it phase changes to steam.
Slower moving water may allow that small volume of water to cool down more, but it will remove less BTU's over time. While that slow water is cooling down more, the water near the heat source is heating up more. But since the boundry layer is now thicker, less heat is transfered to the coolant. Your water coming out of the radiator may be a bit cooler, but both the coolant going in AND the heat source are much hotter.
Now here's the kink in the system which only applies if you compare regular coolant systems to automotive coolant systems. Most non-auto closed systems have a pump that is cooled by the coolant itself. That's when you can get too big of a pump - it can become the main heat source if your original heat source isn't too big. I've done work on coolant setups where 3/10th of a degree can make or break a setup. The coolant itself only gets a few degrees above ambient, but the item to be cooled gets near 200F. It's a mixture of very concentrated heat, and very high flow rates for the BTU's dealt with.