I have a huge laundry list of features I'd like to see, but most of them resemble improvements we've seen in EU4 compared to EU3, or CK2 compared to CK1, so I'm not really worried about that. But I'd like to make a case that the game should 1) try to create a deeper economic and political system, a sort of Victoria for the classical period, and 2) extend the map to India and China to make full use of those new and better mechanics. These same mechanics would also suit timeline extensions backwards (say, to the Greco-Persian Wars) or forwards (to the Crisis of the Third Century or the reign of Justinian).
One, I think the general EU:R pattern of highly-abstracted POPs in enslaved, free, and aristocratic types is a decent approach, but the way that these three POP types affected gameplay (slaves producing tax and nothing else, freemen producing manpower and nothing else, aristocrats producing research and nothing else) was too simplistic and inflexible. It left no room for merchants, artisans, professional soldiers, or clergy, who all made up tiny minorities of the overall population but had an outsized influence on government, politics and the economy. It also wasn't suitable for modeling Roman society in the late Republic or the Principate, with standing armies, a declining 'freeman' base (in the sense of yeoman farmers who filled military obligations to the state), the growth of latifundia, swelling cities and expanding trade and manufacturing networks. The population system needs to be opened up somewhat if it's going to model more than just the pristine city-state society that EU:R's approach suited.
Two, EU:R's trade route mechanic needs to be fleshed out to model both long-distance trade routes (like those for luxuries including amber, glassware, silk and spices) and flows of food surpluses (as in the annona system that sustained the city of Rome and, eventually, a network of second- and third-tier cities that fed into it). I'd like to see this handled in an organic and naturalistic way, with the flows of trade goods and food supplies being pushed or pulled over time by policy decisions or the construction of infrastructure rather than sharply snapping in response to immediate player actions (so, for example, fighting piracy, improving ports and roads, adjusting taxation policies or establishing subsidies in certain regions will shape movement, as will disruptive events like battles, raids and famines).
Three, the map should be extended because key points in the Roman timeline match with important and interesting events in central Asia, China and India. The Punic Wars match up to the high point of the Maurya Empire in India and to the Qin unification and Han revolution in China. In China's case especially, mechanics that would make for strong Rome gameplay would be an excellent fit, especially during the Principate. The annona that supported the large population in Rome was matched by grain shipments that sustained the Han capitol at Chang'an. Both made the transition from mass infantry levies to professional border armies and to foederati at around the same time. Both invested in massive public works to support trade, agriculture, and the movement of armies. The governing ethos (an autocrat at the center surrounded by a professional corps of bureaucrats, judges and military officers) was strikingly similar. Korea and the Tarim basin were to the Han what Dacia and northern Britain were to Rome. Roman colonization in Gaul and Spain roughly match Han expansion south of the Yangzi. And so on. More broadly, in classical times the eastern Mediterranean, central Asia, India and China were already tied together by networks of trade and by events like the Maurya conquest of Bactria and Arachosia, the Parthian conquest of Iran, and the Yuezhi migration from Xinjiang to northern India.
I have a few more concrete things I'd like to see, but I think these three bullets hit the most important aspects.
