I do not know specifically if it kills some bacteria (or other microbes) over others. As a chemist I can tell you that its action is not like antibiotics - it doesn't wiggle its way in and affect some complex biological process. It's more like bleach, a strong oxidizer, a chemical sledgehammer. It oxidizes any organic matter it touches. I suppose some microbes could have more resistant outer membranes. That's the only way it could work. I kinda doubt it. As to the other aspects of the OP that you didn't really ask about...humor me: :-] There is no way it has an ability to 'remove built up salts'. Salts generally can't be oxidized or even solubilized by peroxide. I suppose it COULD help a drowned root system by providing oxygen, which is released as the peroxide reacts with any organic matter it touches, including inanimate organic matter. The anecdote about repotting a plant almost dead from overwatering...well if there was no control plant that was repotted the same with no peroxide added, there's no way to prove the peroxide had any effect whatsoever. That's not even a scientific study with 1 sample, it's nothing.

There seem to be a lot of chemists on this thread, so hopefully one of you can check my thinking here: Peracetic acid is created in large amounts by mixing concentrated H202 with concentrated acetic acid. But in the final dilution, there should be an equilibrium between the three compounds (H202, acetic acid, peracetic acid), in which peracetic acid is only present in small concentrations. However, peracetic acid reacts very rapidly, and the equilibrium provides a constant source of it as it is used up, making it effective even though it's only barely present. If this is case, shouldn't dilute vinegar and dilute hydrogen peroxide provide an almost identical product, if used immediately? The addition of a compatible surfactant, which seems to include common spreader-stickers and SLS, would bring it even closer. As far as I'm aware, acetic acid doesn't have any significant signaling effects on plants. Conveniently, salicylic acid is compatible with hydrogen peroxide - it's used as an aromatic stabilizer in cleaning formulas. I've long used hydrogen peroxide as a foliar "disinfectant" for ectoparasitic fungi and some bacterial diseases. It works - the main downside is the frequency of application, which is reflected in some of the studies linked above. Even full strength drugstore peroxide, 3%, doesn't seem to harm most plants as long as a small amount of spreader-sticker or surfactant is included to keep the drops from collecting/concentrated in small areas. It's pretty much equivalent to dilute bleach in this application, but without the hazards of applying sodium and chloride ions to leaves. I can't see the full paper linked in the original post - for those who have read it, how big was the difference between the treated and control? Often flat effectiveness is reported even when the actual effect size is very small.

Doubt you even need to try the peroxide. For one thing what are the "centipede type things" you have? Size, color, damage if any? And when it comes to fungus gnats, pull back on how you water. The simplest thing to do is to allow the soil surface to dry. But to get a running start on stopping the gnat's life cycle, use Bacillus thuringiensis israelensis (either Mosquito Dunks or the granular form) in the water you give to your plants. Realize that it takes several weeks to stop the life cycle. And those "unknown yellow round clusters of eggs" may well be Osmocote, a slow-release fertilizer.

I bought dunks and am trying them out. Boy, I hope this works!