There is many software synthesizer providing controls for both "voices" and "chorus" effect. I guess that both works by adding more oscillators layers with little differences, but what exactly change between the two functions?
Voices are separate oscillators, chorus is merely a delay-based effect.
Usually the number of voices in a synth actually refers to the number of oscillator groups. For example, an analog synth with "six voices, each with two oscillators" means that you can separately program two different oscillator programs (e.g., wave shape, frequency, etc.), and then play six different notes at the same time, each note playing two oscillator programs. So there are a total of twelve oscillators, but you can only program them in banks of six.
Chorus is a delay based effect where whatever signal comes in is copied and delayed slightly, usually pitch-shifted also, and then the copy is played back with the original. The delay and pitch changes are usually varied over time to make it sound like more than one instrument is playing the same part.
'Chorus' is a similar effect because it is also achieved by adding similar voices together. It is technically correct so long as the voices still create a sound that is perceived as a 'chorus' and not as distinct voices or pitches:
Sometimes you can have oscillator voices synchronized in Unison mode, or otherwise have their phase playback "freely". In the latter case, detuning the voices creates a stronger chorus-like effect, due to the various "beating" introduced by superimposition of the waves at various phases. This is similar to introducing one or more delays. When the Osc phase is put in "sync" mode, generally a sharper flanger-like effect is heard, as the voices cycle in and out of phase uniformly, from the master oscillator.