I'm definitely going to move up to 16-bit sounds eventually. The combination of low-res samples and retro melodic sound design has always made the SNES my favorite sounding system, and FM is one of those daunting, tough-to-crack types of synthesis that I want to explore in depth. The article I'm working on at the moment is actually about making seamless loops, which will be important if I'm going to talk about SNES sounds.
I started with NES games because the simplicity of the 2A03 makes a thorough analysis possible, and lets me develop my technique for actually transcribing the individual sounds. The Pacman sounds are the first I've done that weren't based on a very simple set of waveforms, and it was significantly more difficult and time-consuming (though also more rewarding and educational!) Finding ways to mimic the samples used in SNES games or to reverse engineer FM patches for Sega games is going to be a challenge, but I'm excited to see how far I can get with it.
And I totally agree that there was a very long lazy period of game sound design. As the industry's gotten bigger and as cheaper technologies and accessible learning resources have emerged, the average level of sound design skill behind development has increased. Hopefully, I can add something to the growing pool of educational material.