Well, you need to individually select the instrument per channel in TTS... if you don't, then everything will play as a piano.
Why not in the first instance keep using the motif as a sound module, driven from Sonar?.. then, change out instruments one at a time to suit...