Wonderful game. It's very rare for projects so early in the dev cycle to be this solid and functional, and it's good to see that you've decided to continue developing this one further.
My main piece of advice is this: compartmentalize. Make further additions to the game modular, avoid deep and irreversible changes to the core formula. For example, as the game is right now, it can extend a single 'playthrough' of itself forever; with a save system, someone could have a 20k+ tile map months or years down the line, and keep expanding on it up until the game cannot physically handle the map any more (which is something that could be fixed or worked on, too). And while something like that lacks a solid/satisfying endgame and doesn't truly appeal to many people, I feel like it would be important to preserve it to some extent, as it fills a very particular niche that some people will really appreciate.
Modularization of the content and mechanics expansions would allow something like that to remain, while still allowing you to change and evolve the game from the core up if desired. So for instance, you could have the option to turn on or off settings/content that directly or indirectly establish an end scenario or win/lose condition, which changes the entire gameplay experience, without it affecting other aspects like the tile variety, added mechanics, or other changes to the experience that can all work separately of whether the game is a potentially infinite relaxing builder, a super complicated tile management puzzle, or something in between.
The difficulty with this is that it requires a noticeable amount of extra work, planning and testing. Making sure all the 'pieces' fit together in different configurations, that they work well when mixed all together and when used just by themselves along the core, determining which ones need to be used in conjunction or are mutually exclusive... It's not easy at all, but I believe it's worth the effort. A smaller scale would help if you decide to go through, taking smaller steps will mean every one is more manageable. Hell, that last piece of advice applies to any project, whether it's something heavily compartmentalized or not.
One last little detail, as someone with a 1000+ tile game still going; some people have mentioned that building long stretches of tiles that aren't very interconnected allows you to metagame and handle certain missions very easily. It's an issue at surface level, as when you get further into the game you can effectively turn those "ugly" stretches of tiles into a challenge of making them look good; thus, a solution to a short term problem becomes a long term problem, which I find an interesting and fun gameplay mechanic, personally. The only thing that ends up looking a bit aesthetically 'wrong' is the small, often single-tile patches of farmland that you have to place down to progress certain missions; I feel like needing a house tile adjacent to a field group for it to count as a farm (which is what I end up doing a lot of the time) would fix it, making it look like there's plenty of little family farms scattered through the land, instead of random splotches of fields in the middle of nowhere. The 'village' groups could similarly require a minimum number of houses to qualify, though I like the look of the scattered houses around the countryside, when they're grouped together they remind me of villages that are just a loose collection of nearby farms and cottages.
Anyways, you've done a great work with this game, and I hope you keep working on it and it ends up doing well :)
Unless they've changed it at some point, tiles where rotating is relevant, which is just rail tiles, already auto-rotate every time you select/reselect a tile space. So if you have a rail tile with 3 connections and you want it in a Y shape, all you have to do is repeatedly mouse over the tile you want to place it in until it shows the shape you want.
