Starting with a relatively simple subtractive software synth that uses primitive "analog" waveforms -- like, say, Tal-Uno-62 (https://tal-software.com/products/tal-u-no-62) -- is probably the best way to play around with learning synthesis. It's a lot harder to create a "bad" sound in a simple subtractive synth, and in something like Tal-Uno-62, there aren't a whole lot of possibly confusing menu options and buttons.
Samples and wavetable synthesis are also relatively easy to get a "good" sound of quickly. Most are structured, architecture wise like a subtractive synthesizer, you just replace the primitive waveform with either a sample, or a bank of digital waveforms.
FM and additive (and spectral and granular) are neat techniques, but IMHO way more finicky to program. (They are also techniques where the software interface is quite welcome! Programming a DX sound in the (free and open source) Dexed VST is quite a bit easier to do than it is on a Yamaha DX7's two-line LCD screen and buttons.)
For the parent poster: I understand that quite a few professionals, particularly in loop / beat oriented music genres (such as hip hop / trap, EDM, and reggaeton) use FL Studio quite a bit! The nice thing about FL Studio is that it has a nice intuitive layout for beat-making. It's what I'd personally recommend for someone just wanting to mess with synthesizer beats. (My personal favorite is Reaper but I don't tend to write beat oriented stuff. :) )
Interesting, for me additive is the most intuitive. I find it easier to visualize sums of sine waves than subtracting harmonics from a more complex waveform (such as a sawtooth), particularly when considering the use of notch or bandpass filters. But that may have a lot to do with my education, which featured a heavy dose of Fourier Transforms and associated techniques.
On the other hand, "simple" additive with relatively few partials is actually pretty easy too. (The Hammond B3 and other tonewheel organs are essentially additive synthesizers.) A lot of software (your Alchemys and Razors) is built on a simplified version of additive too, or uses additive techniques to perform "resynthesis", or has other tools that "simplify" some of the additive editing.
The version history on KVR is interesting: https://www.kvraudio.com/product/circle-by-future-audio-work...
Helm's first release was August 3, 2015: https://github.com/mtytel/helm/releases
Circle's first release was also August 3, 2015. Maybe they're developed by the same person.
Here's a rough crash course from someone who also struggled to get started:
1: You start out with a wave (sine, saw, pulse, etc). No synth starts out with a pure form of these waves. They all add a little color. Most synths have 2-5 oscillators (they produce the waves), often with each having different controls.
2: The filter constrains the sound. So, for example, if you wanted to make a bassy instrument, you'd filter out most of the high frequencies. The Attack, Sustain, Delay, and Release settings affect length and volume over time. This is what you use to turn waves into plucks, kicks, claps, strings, and anything else you can think of.
3: Most synths have at least 2 LFOs, or low frequency oscillators. You can connect these to different controls across the synth and change the timbre in weird, sometimes unpredictable and amazing ways. With Helm in particular, you click the little helmet by the LFO and drag the little bars on the different controls to change how much the LFO affects it. You'll see a little green bar pop out and move with the LFO. Serum has a separate tab with LFOs and routing to synth controls.