...permit me to voice a minority opinion. IMO you are likely to waste years trying to learn and apply TA, only learn that most of it is useless. I recommend that you subscribe to a data/charting/backtesting service appropriate to the time frame you wish to trade. Then learn how to recognize pure price/time patterns and code them into trading algorithms. Backtest them, and trade the best one you find. This may sound like a daunting task, but if you go down the TA path in retrospect it will seem to you that the time you wasted "learning" TA would have been better invested in recognizing profitable pure price patterns.
You're equating "TA" with indicators while it encompasses everything you include in your post: price action in the context of demand/supply, price patterns, and indicators. It's not necessary to subscribe to anything in order to learn pattern, if that's what he wants to do. Just pick up a copy of Schabacker. Indicators can be very useful, but they are only part of the puzzle. If an individual expects them to be sure-fire money-makers, that's hardly the fault of the indicators.
...as I said, it's merely an opinion. I agree with you about the Schabacker, a great book to start with. And the Achelis (especially the published book) is very thorough. Forgive me if I overgeneralized about TA, but using Achelis as an example, I would guess that over 90% of what's in there is not simple Schabacker-like pure price patterns, but rather indicators, oscillators, sentiment, ratios, et al. Also, where I am coming from is intraday trading on one-minute charts, and I have tested most Schabacker-like patterns and found them wanting. IMO the best use of such patterns is in monthly charts.
You're correct that Achelis is mostly indicators. But then I didn't say it wasn't. The original question had to do with indicators, and Achelis is one of the more complete encyclopedias. The "right" way to study TA? First, understand that it includes all of what I listed. Second, begin at the beginning, with price action, which means going back to Rhea and Wyckoff. Then, pattern, i.e., Schabacker. Then, indicators, i.e., Achelis, Murphy, Elder, Appel, Wilder, Lane, etc. Most people, however, aren't going to want to do all that. In fact, I've known only two or three over the years who've bothered. Everyone else wants to cut to indicators, and nobody elected me TA-police, so if that's where they want to start, okay by me