One reason to record a mono instrument like a guitar in stereo is so that you can later adjust the timber of the instrument via levels. I usually record an acoustic guitar either X-Y between the 2 hands of the player (look up X_Y microphone placement on google) or spread, with one aimed at the "pegs" and one aimed at the 12th fret of the guitar. Fret side is lighter and airier, the "peg" end is deeper and crisper on attack. Then you can mix the two via levels, compression and pan to get the sound you want.
Stereo micing a voice is of little benefit, but recording with two mics set different places can give you options for later mixing with the proximity effect of close micing or the airy sound of being further away. Just be careful if you try to use both, combining the two signals can cause some nasty comb filters that make some frequencies disappear while others will be boosted and make mixing consistent levels near impossible at times.
As for setting it up in Sonar, I usually use 2 mono tracks so I have maximum flexibility. If you record to one stereo track, you can only process both sides the same. You do have the option of "split to mono" but their is no real gain to not doing it this way to start with.
Understand though, a single voice and a guitar are both "mono instruments" with the spacing of a guitars strings making a stereo recording sometimes advantageous. A single voice does not really benefit at all from stereo except for every live environments where you are actually recording the room in stereo more than the voice itself.