Limiting individual tracks to gain headroom on the master does not work. Frequency buildups generate new peaks.
Limiters on tracks have rather the use to let the track sit better in the mix. Like compressors, they guarantee a dynamically stable signal or can be used for depth staggering. Less transients = further away.

 

I agree, but I use a limiter to protect my ears, and also as an early warning system to notify me if the mix is getting too hot. It doesn't do any actual limiting on the final mix, it just keeps me disciplined.

 

Using compressors, saturation, clippers, group processing etc on tracks ultimately has one purpose, so your limiter at the very end of the masterbus chain doesn't have to do a lot of work, which means, less artifact. Adding limiter on every track means you are adding more and more artifacts and I don't think it will result in a good sounding mix.

If you ever need a limiter on every single track, it can sometimes means you aren't mixing at proper sound pressure monitoring environment. Boost that monitor volume and pull every fader down so you gain more headroom so everything you hear is just the same as before but you pulled all the peaks away from the red.

If it is just few tracks here and there, it's okay I guess, I sometimes add limiter (or clipper) to 808 (not drums, the 808 bassline) track just to be safe.

 

Channel treatment should only serve one purpose: sound.

 

I work mainly with acoustic based instruments and not in only electronic so with electronic I cannot comment. Acoustics I can.

With acoustic instruments, if they are recorded properly they have a fabulous dynamic range. While the most obvious transient, frequency cancellation comments could enter and you'd be right; it is desirable to keep as much of the natural dynamic range and performance dynamics as possible. Automated levelling or forced ceilings kill performance dynamics. There are people like
No Avenger and others here that can point you to why compression is a better choice even though I know why because they can validate it not just in practical terms, they can show the industry and theoretical reasons why it works in practice - and when to choose upwards compression instead of limiting.

If you want a tune to be all the identical volume with zero performance dynamics then perhaps your idea might work. I have never tried it or done it and I accept I may be wrong, but it automatically looks like overkill when putting it on a bus or the stereo out in mastering would do the job just as well. I tend to not limit if I do not have to as it seems to work better at the mastering stage. each to their own.