A neuron in Catacomb 2 is simply a compound assembly which has been assigned the property of being a neuron. That is, calling things neurons adds conceptual information to a model, but does not change what gets computed. Note, however, that future versions may use this information to try various optimisations which may be applicable to neurons.