What is very different though is the use of a 63A B-curve MCB as input isolator.
Certainly I think isolation should always be all-pole (though for TPN that is not so common and ideally then you want a mate first / break last on N) and I was surprised that ROI did not do so in the past.
The use of the MCB to me is kind of mixed, I can see it is a good idea to limit each CU's total current without pushing the DNO fuse in to that job (I presume ROI still has the equivalent of a fuse as ultimate limiter?) but it also seems a bad idea on selectivity as many high current faults will take out both the faulty circuit and the whole damned supply!
63A B-curve instant trip is 189-315A, though looking at the Hager tables it would be selective with a 20A B-curve for socket radial to around 700A. That is still less than often seen for PSSC on sockets here (my experience mostly the RFC, maybe easier at ends of 20A radial though?).