Your question highlights CDP's rather piecemeal approach to mixing. Mixfiles are not the answer: the syntax gives each sound a level, pan position and start-time, but you cannot stop the sound early or start part-way into the file. Extend Sequence 2 is quite like a mix and gives each sound an overall duration, so it can be stopped early, but not started late. It specifies level, but not pan-position.
Submix Merge mixes only two sounds, but you can specify a stagger value (soundfile2 enters stagger seconds after soundfile 1) and/or a skip time (start mixing sound 2 at skip seconds into the file). You can also specify overall start and end times, but not the end time for each sound. The function has no level controls. Conversely, Submix Faders has a time-varying level control for each sound as well as an overall envelope for the mix, so this could be used to fade sounds in and out, but it has no control over individual start times.
Finally, Soundshaper has a Premix function, which is a composite of pan, level and start-time (which prefixes silence). It can be applied to each sound before mixing with e.g. Submix Mergemany. Currently, it only deals with the whole soundfile, but it might be possible to develop further controls in the future.
