What if Voynich plaintext is very heavily shortened?

[Koen G]

The idea of tokens formed by common groups of glyphs is a very natural one, however it can be observed that their frequency is extremely inconsistent, ranging from 0 to a lot, and therefore difficult to reconcile with a simple interpretation as a plaintext letter or group of letters.

Voynichese as a whole is not a consistent system though, so wouldn't this argument work against most cipher-based solutions? We're always dealing with one subsystem shifting into another, or certain preferences dominating different sections.