There are two parameters that control detune, and that depend on each other. Depth and Rate.
The chorus applies no fixed detune, but will instead “waver” a delay of the sound forth and back with an LFO, first consecutively delaying/lagging it and thus slowing it down, then speeding it up again to catch up. While it is lagged, it will detune lower than the original sound, yet while it is then catching up again, it will detune the sound to have higher pitch than original. So it is not like a pitch shifter, it can only detune down to then detune up again, always alternating these states with the speed of the LFO.
So it will alternate between detune up and down. The depth is the amount of maximum delay and thus the strength of the detune up/down, and also the detune will depend on the speed, i.e. if the sound is wavered faster, it will have stronger detune as a result.
stacking the chorus devices can make them sound a bit like an ensemble effect. A bit different, but it is possible to emulate. Each chorus device will add a detuned voice so to say. Even more, because it will detune not only the dry signal, but also the wet of the previous chorus devices, so take care not to end up in a big mud too soon.
You should choose a dry/wet ratio like only 25-50% wet, so the next chorus device will also manipulate the dry signal.
Also the rate can get tricky to choose so the detune is always spread nicely and not generate hickups. I.e. choose one rate, then for the next chorus divide the LFO frequency by 15 and mul by 14, then every 15 cycles of the first chorus the second will make 14 cycles in the same time. Generally try to give the chorus devices different rates to get an ensemble like sound, if the rates all match the sound will be too much in sync to sound like a multiplication of voices.
You can also try to use the chorus filters to make it affect different bands of the sound, so the mids get different wavering than the highs.
Using the “delay” parameter you can offset the wavering sound in time, thus also creating a stronger illusion of multiple voices. For a stack of chorus devices you could use different delay for each stage (like with 15ms steps for each instance), so each chorus voice has a different phase in comparison to the original sound.