As David Hess, it is only useful for certain applications where you need to know the true average Voltage of a non Sinusoid waveform.
Poor use of the word "average", I'm afraid. The mathematic purists would be apoplectic.
RMS is a mathematical function. Applied to a voltage, it takes any given waveform and calculates the equivalent DC voltage that would produce the same power dissipation in a resistive load. For a sine wave, this calculation is very easy - the RMS voltage is 0.707 times the peak voltage. As a mathematical function, it can be applied to any quantity measured against another. Voltage (instantaneous) against time is the classic, but instantaneous current against time is just as valid.
Because sine waves are common enough, the cheap non true RMS meters have this calculation effectively built in. Measure a sine wave with them and you'll get the right answer. However, measure a non-sine wave and you will be shown a lie. The less the waveform looks like a sine wave, the bigger the lie.
A true RMS meter pays attention to the actual waveform - not just the peak - and will give you an accurate measurement for a wide variety of waveforms other than sinusoidal.
Note: There are limitations that affect even true RMS meters. Check out "crest factor" if you want to get a full understanding.
Most of the time, AC volts is just used to check mains voltage is there and is it the right number of volts.
Or maybe checking the secondary side of transformers.
I have never had any reason to need the rms feature of any of my dmms.
(Maybe I just do not do enough electronics or the wrong type of electronics )
If any of your meters have true RMS, you've almost certainly been using it.