You're correct, but this is a mostly academic distinction on a network like Hong Kong's. The gap between trains is a much more important human metric, and it changes based on where you are in the network.

An example: you could fill as many trains as you could physically fit on a track coming from Central (the CBD) during rush hour; a train every 30 seconds or so is crucial to keeping things civil on the platform. As you get farther from Central, utilization falls until you reach a terminal station, where the time between trains can be more like 2-3 minutes without any problems. Through clever signaling you can achieve a much higher perceived throughput on the busiest stretches without actually running more trains.