Well that is a good point. But it depends on the design really. I remember reading about a pebble bed reactor which used liquid hydrogen as the coolant which was a lot more dense than this reactor should be for example. Problem is hydrogen likes to erode the carbon typically used to cover the fuel pellets leading to radiation leaks. Oops. I suppose they could use liquid helium but that is even harder to keep at temperature.
This is why I say something kind of like the ACP100/S design with integrated components inside the same casing would make a lot more sense. Only thing that design does not have is high fuel lifetime but they can handle that by changing it to use higher grade fuel or more of it.