That depend on how you understand voices, and on what you wan't to do.
Voices can for example be detuned differently. So in a RANDOM mode you can get a random detuning for each note you play. Let's for example say, that you play the same note repeatedly, then you will get randomly variations on that one note each time you press the key. I would like the detune was a random amount of detune, so it actually emulated the behavior of the original CS-80. CS-80 v2 actually had such a global random detuning in a set area.
Voices also have other parameters that can be set individually for each voice.
Playing notes that use those voices with a seperate setting for each randomly thus create random variations in the sound. Why won't you have this?
When you say "simultaneously", then to me it for example could mean playing many notes that use one voice each. The random playmode does'nt change this, but adopt the random behavior for each note that in this case also is for each voice as one note in this case equals one voice.
"Simultaneously" can also be understood as UNISON, where 2 or more voices play at the same time for a note. If you wan't that to be in a monophonic/ single note sound, then you would select a UNISON voice mode. In this case you does'nt use the Random mode.
As you for example also have an option to assign one of up to 4 different sounds to each voice, then the RANDOM voice mode can also create small or huge sound differences in each note you play. And it can also be used to create speciel effects in any preset.
A simple example of a single possibility with the RANDOM voice mode.
Select the template preset "Multi Sequence". Press and hold a note and listen to how it play with it's Z1 voice mode "ROTATE".
Then shift the Z1 voice mode to "RANDOM". Press and hold a note and listen again.
Turn off the arpeggiator and play a single note repeatedly both in ROTATE and in RANDOM voice mode and listen.
Another example preset could be the template preset "4 Voices rotate" that for the understanding perhaps should have been named "4 sound rotate" or even more accurate using the matrix terms "4 Single patch rotate".
Play the sound in different ways while you listen. Perhaps use only 1 note at first.
Then change the Z1 voice setting to RANDOM - play the same and listen again.
The try to change the second column "Zone" to Z2, so you have the left column having Single patch 1-4 on Z1 and the right column the same using Z2. The try different Voice mode settings for Z1 and Z2 and listen.
Also try to have different voice modes for each of the zones.
Do your own experiments too with parameter variations.
Perhaps try other voice modes with differnt presets including with sounds that have long release time to listen to the differences between the voice modes.
I have to ask you - what are your definition of a voice? I don't understand why you should'nt want this original RANDOM CS-80 voicemode.