Why do it? Because mounting many modules is much quicker than activating during boot -> it can be more smaller base modules -> easier repack and updatelooks bit complicated - maybe you could do some work in this direction?
My example for Firefox:
1. Create to-mount and firefox folders in optional (/mnt/live/porteus/optional/to-mount/firefox)
2. Move your 04-firefox.xzm module from base to optional/to-mount/firefox
3. Download firefox-[mount]-$arch-1tm.xzm (~0 MB) to base https://www.dropbox.com/sh/ui4ydoh3j34x ... C%20noarch
4. Reboot without any load= cheatcode (required).
Because Porteus releases are/were in frozen state (base modules are not updated) and probably no one uses changes folder for firefox updates and similar therefore mounting read-only can replace activating (maybe if one file = one link then updates will work but to much unneeded work). I use this solution, and content of firefox-[mount]-$arch-1tm.xzm is merged with another little modules. If default software packages (deb, tar.xz etc) can be mounted read-only, it would be great and no need to convert to xzm.