I am surprised that the community here is so hostile. The OP just asked for some answers to a few questions. It's fun to talk about your passion, hopefully with other people who are passionate about the same things. Thats what I see in this thread. So, to that accord, I will suggest these challenges:

• Try and scale this concept into something completable in 90 days or less.

As far as chronological order, here is my advice (taking into account the above suggestion)

1. Platforms (Distribution)
2. Features/Mechanics
3. Technical options for features related to your engine of choice
4. Prototype
5. Art
6. Test
7. Polish

Good luck!

I think a single person in today's modern world could create something along the lines of FF3/FFVI. Honestly, the easy way to do it would be to use RPGMaker VX Ace. I challenge you to not do that. I don't expect you to use C++. I'd say to use Construct 2 since that seems to be what you like, but I'm not sure if Construct can handle it. It isn't the gameplay I worry about, rather I don't remember seeing anything about data structures. If Construct 2 has a way to store a form of database for the items, weapons, enemies, etc... than maybe you can do it. Otherwise, I'd recommend GameMaker, as I know for sure it has the data structures, etc... to handle a massive RPG like this.

As far as being alone and able to do this....I'm not saying you'll be able to do it in a short time like a month or two, but yes, you could do it. It might take you a year or more to fully flesh out the story and everything else, but I'm sure you could get at least the concept going within a couple of months, with at least the beginning of the story, the mechanics, and a small area and only a few enemies. Then, it is a matter of time creating all of the enemies, weapons, items, etc... then you need the places to go, which include the graphics, then the design(if you are doing it via tile engine), and finally the other things for the places, like enemies, treasures, quests, etc...

The order of design....I'd probably do it similar to what I've said in the above paragraph. First, I'd get at least part of the design ready, especially the core concepts, like style, battle types, etc... This is because if there are any special battle types for example(from behind, caught in the middle, enemy caught in the middle) or anything else special about mechanics(like critical hits, changing battle scenery, speedup/slowdown of ATB, almost anything outside of normal), you need to account for it. It's like if you are going to have general speedup, you'd need to put a variable somewhere that the ATB accesses. If you were making a fighting game, you could design for a single player...but instead you would design the fighters the same, and then have be able to be controlled by Player 1, Player 2, or CPU. You would design this from the start, not after you have already hard coded the fighter on the left to use Player 1's controls. This would help you avoid changing things later that you've already done.

Then, I'd use a bit of programmer art, and get the engine itself working. I'd get the tile engine, fighting, treasures, weapons, shields, everything else you can equip ready. This also includes all fight mechanics, including magic, special abilities, things like espers(FF3/6), whatever you are going to need. Don't worry about actual content, rather get things ready for content. Make the party system ready to have 4 characters, even if they are simple blobs for the moment, and make it ready to change characters as you need it to. Make the characters have stats, etc... make them be able to equip things. Make the store system work to buy items and weapons. Get the item/weapon/magic/everything else database started and working, so you can add things as needed. Make the AI system etc... for enemies too, even though they are just blobs. Basically, get the base engine done, taking into account the design choices you made.

Next, you can crack down the actual content. You need to make your final decisions as to what that content will be, who the characters are, where they go, what they do, and what happens to them. There are 3 parts to this stage, and they can be done in bits as you go. I just wrote about design, which includes the story and final decisions about these things. Then you get programming and art. You need to design the art(including sounds/music) for all of the locations, and then you need to program said art and locations. I mean to say that you need to create the grass tiles, and you need to map out the maps using said grass tiles, etc... This "programming" part is also where you would place treasures, etc... and add items to the database to link the treasure boxes to. As you are implementing the map designs, you also configure enemy random battles using the base system you hopefully did in the paragraph before as it is a core mechanic.

After you get all of these things done, you can run a final polish stage. This could include adding things you didn't have done yet, like particle effects for magic, balance and tweak things like level up speeds and strengths of enemies(stats too). Maybe you need to work on some quest branching?? Get that last 10%(or 90% as they say) done.

You'll need a testing stage somewhere in here, although much of the testing was likely done before you started adding all of the content, but in any case, you need to test all the content to, and the final polish as well.

Final Step...wait for it..................PROFIT!!!