I would have said the NES is the least worrisome platform, give or take some of the FDS stuff, until you hit things like chip8. After that it is SNES, anything http://problemkaputt.de/
has tackled, MAME/MESS and the older "home computers" and then it gets fun.
On a "tree" then I am not sure how much different it would ultimately be to what http://www.romhacking.net/start/
"If we could develop tools that could make rom hacking seem less overwhelming"
Tile editors are about as basic as they are going to get*, I am not seeing how assembly could get much easier**, atlas and cartographer (or kruptar and oriton) are as complex as they are by necessity ( http://transcorp.romhacking.net/scratchpad/Table%20File%20Format.txt
makes as convincing a case as any), a hex editor is a hex editor and as we have had to say on many occasions are hardly elegant tools but brutally simple things without much nuance to them. This leaves getting debuggers up to scratch (we have had high level discussions before but "like fceux but for ?" is what it usually boils down to or going down the path of pokemon hacking, which is to say lots of tools but not much for those wanting to learn hacking.
"We're using tools designed in the early 2000s" and it shows in the UI and languages used but I am not seeing the great issues/new things afforded by improved hardware beyond being able to emulate newer systems -- even something as powerful as IDA we largely saw in X86 at least in the late 1980's if you left debug flags in. It is nice being able to do a find and replace in a fraction of a second or a full relinking search but they are niceties on the high end rather than lowering the bar type changes.
Several have considered some kind of code fragments written in python or something to help people with building their own game specific tools but this is perhaps a slightly different matter.
*there is a tile viewer called tileggd which eschews console naming conventions (NES 2bpp, GBA 8BPP....) in favour of more graphics and code focused approaches (in the end there are only so many ways to order alpha, define the number of bits per pixel and whether said bits are in big or little endian after all) and gains a tiny bit of flexibility in the process.
**I guess if someone wants to develop GBAATM and DSATM style cheat injection tools for older platforms but with a more ROM hacking twist it could help, not by much though. ROM hacking is hardly a good candidate for HLA style ( http://www.plantation-productions.com/Webster/HighLevelAsm/index.html
) "libraries" either.
"a couple of videos"
I like video, I like film making, I would consider myself reasonably proficient at editing video and I also like ROM hacking and would consider myself reasonably proficient there too. Despite thinking about it on several occasions now I am at a complete loss on how to make a worthwhile hacking tutorial video though, save for things I could probably summarise in a paragraph. If someone can crack that though (choice video https://www.youtube.com/watch?v=uFfq2zblGXw
) then I am there.
More generally then sure you hear stories of people discovering emulation or people discovering translated games despite using emulators/hacked devices for years but the awareness among those sets is not bad, even if ROM hacking= user driven translation/localisation more often than might be ideal. Raising a bit of awareness there is not a bad thing and https://www.youtube.com/watch?v=PmLi8Mx7Yf0
makes me cringe every time* (some seem to like it though) but is one of the more popular examples of non translation hacking out there.
*I figure it must be like what the youtube safety patrol feel when they see various machining and woodwork channels.