Importantly, according to this https://physics.stackexchange.com/a/191465/7911 https://physics.stackexchange.com/a/191465/7911 the efficiency of steam is not

Importantly, according to this https://physics.stackexchange.com/a/191465/7911 https://physics.stackexchange.com/a/191465/7911 the efficiency of steam is not far from the thermodynamical maximum under the assumption that the nuclear energy is initially converted into thermal heat in solid objects. You can do better, but only by harnessing the fission products (neutrons, protons, etc.) themselves before they thermalizes with surrounding material. This is possible, but it's a completely new and separate challenge in nuclear engineering, with a maximum improvement of about a factor of 2 in energy efficiency.