A few basic problems which we can explore in more detail:
- Finances. Everyone simply has more money in the game than Early Modern governments tended to in reality. It would be unreasonable to expect Paradox to represent the full depth and breadth of differences in taxation and spending between different states, but suffice it to say that both armies and navies don't cost enough. This is tricky though, because unless it's designed well it could be less fun to have to deal with cost rather than more. One potential way to deal with it would be to go the 'Sweden' route and let armies loot territory they're in in exchange for costing less maintenance - bellum se ipsum alet.
- The state army is the only army. Right now the only way to have an army is to utilize state resources, state money, and state manpower. If your country runs out of those, you cannot fight. This isn't historical. Despite the centralization of power in the Early Modern Era, non-state actors could still produce military forces of immense strength (and lots of money, too!). This ties into the issue above. Think of Poland-Lithuania during the Deluge: the Commonwealth's army was easily defeated in 1655/6 and almost the entire country was overrun by Sweden, Russia, and the Cossack State, yet before too long they were able to recover. Sweden's policy of making war pay for itself, its soldiers' disrespect for Catholicism, and failure to appease the Polish nobility led to mass insurrections in the countryside as the magnates used their large retinues and great wealth to combat the Swedish occupation, eventually enabling the king to return from exile. This isn't really covered by the game's 'rebels'.
- The AI and peace. Thanks to Jomini for bringing up this issue: The AI's reluctance to make a quick peace, both when losing and winning, results in wars of total occupation occurring very frequently because the AI is reluctant to give up provinces until it's had much of the country occupied. By that time it is usually weak enough that it is desirable to occupy even more in order to get a greater peace deal, and so the spiral continues. I'm sure he can explain it more eloquently than I can.
- The relationship between fortresses and field armies. I started a thread a while ago on this topic, and it's which I think is particularly important. I don't still hold to 100% of the views I put forth in the linked thread, but most of them are still relevant. Mainly, that the current system in which armies have no means at all to avoid enemy armies if they can't get out of a province in time is unrealistic to an extreme degree, and that letting them garrison fortresses is one possible solution to that problem (although not a particularly historical one, but I see few better alternatives). Also, that fortresses should have a means to fight back against besieging armies. Currently they're completely passive during sieges, which could do with changing. Siege warfare in general needs to have more depth. For most of the time period it was siege warfare, rather than field battles, which made up the bulk of major military activity. For the period of warfare between Austrian Habsburgs and Ottomans between 1526 and 1683 there were countless sieges and raids launched back and forth, and a whopping two major battles, Mezőkeresztes in 1596 and Szentgotthárd in 1664. In the game battles are ahistorically ubiquitous.
And finally, the one simple change I would make to the military system which I post in every thread like this and which annoys me every time I play: stop giving defensive bonuses to besieging armies!
Unless the game is a perfect copy of what happened in real history, and following that timeline step by step, no matter what happens, no matter what the devs do, you'll always be able to find examples where their systems used in EU4 aren't sufficent/can't properly emulate what you want. This is the nature of a game, and it's suprising to see so many people confused by it. After all, EU4 isn't a history simulator, but a just a game with historical elements to it.