Harris Systems, going back to the EDACS Systems in the GE and Ericsson eras, can use ANY frequency pair as a control channel. Rotating the control channel on a regular basis is considered a good management practice, as it spreads out the 'wear and tear' on the system components over several repeater stations, rather than just one or two repeater stations. It lowers the MTBF (mean time between failures).
Motorola for some reason never did this, and usually only confined their control channels to the first few channels in a frequency set.
Limitation of their technology? ...I don't know.
I can speak on Harris systems, as I designed and installed them...
In a P-25 system (and older EDACS too), there will only ever be a single control channel at a time.
It does not matter if it is a single site system, or a 20 site simulcast cell!
Trying to scan across several systems and control channels is flawed from the start. At 9600 baud data rates, the WORKING CHANNEL ASSIGNMENT message to user radios is on average only 30 milliseconds in length. Miss it, and you have to wait for the message to be repeated on the control channel, at which time you late-enter into the group call.
The only reliable method to listen to several trunked systems, REGARDLESS OF THE TECHNOLOGY FORMAT (EDACS, LTR, Moto, TETRA, MPT-1327, etc.), is to have multiple scanners listening to the systems, one dedicated to each system.
My humble 2¢ worth....