I very much understand what you're saying, from a realism point of view. But I think that the only justification for a combat stat's existence is that it does something to make combat more strategically interesting (and by an amount that is at least commensurate with the amount of complexity it's adding).
If you're placing a lot of effort into creating some kind of rock-scissors-paper cycle where (for example) status effects > defense > armor > status effects, and you can implement it effectively (in the sense of players need to, and can, make these decisions in building their team, and have a way to use it effectively against enemies), then making this kind of split between Armor and Defense could be smart. The added depth is worth the extra complexity.
On the other hand, in your standard Final Fantasy type of setup, all you'd be doing is confusing your player with two stats that essentially do the same thing. They are going to wonder what the difference between the two is, and whether +8 armor is better than +6 defense, and even if you straight-out tell them the answers, they're still going to have to keep an extra non-intuitive element in mind each time they go to the equip screen. This decision isn't any more strategically interesting, it's just more difficult to clearly make; the added depth isn't worth the extra complexity.
In a lot of cases, I'd recommend paring down stats from the basic formula, rather than adding them. Civilization IV is one where I think went a bit too far but it's still a good case study in good TBS combat design, in some ways. In Civ 4, units literally had a single combat stat - "unit strength". An Archer had 3 strength, a Crossbowman 6, and an Artillery 18. As they were damaged they would lose strength, weakening them in every way (and could eventually heal back to their max strength).
Units in Civ 4 also had different abilities, however - Archers gained a big bonus defending cities, Crossbowmen could decimate melee units, and Artillery could break down a city's walls and do collateral damage. In addition, units gained levels and at each one you'd choose a "Promotion", such as "moves faster through rough terrain", "heals units in the same tile", or "immune to collateral damage". These abilities and promotions allowed you to build armies appropriate to your situation, and while there was a lot going on with them, they could get away with it because (most of) the abilities were clear in their strategic use and the single combat stat (unit strength) meant that it was very easy to judge how something like "+50% strength vs. melee units" would affect the odds of the battle. The extreme simplicity of combat stats allowed the system to use very colorful and diverse abilities while keeping things mostly clear for the player.
So, that's a lot of text covering my thoughts on the subject, but to get right down to your question, my instinct is that (based on other things you've posted about your game) you probably shouldn't split the Armor and Defense stats - having a percentage damage reduction in addition to a flat damage reduction probably isn't an interesting enough decision to justify the added complexity that comes from two similar combat stats. I can only think of a small handful of games where I think this kind of thing would be good.
