But I think you're over-estimating how hard it would be to get rid of traffic in many other urban cores in the US. You don't need to do the entire city; you can just do the densest parts. It's not that hard to beef up a city's existing bus services and use them to totally replace cars.
There's been a lot more reclaimed land than that added to Manhattan since 1811, but they mostly slotted into the pre-existing established grid system.
But the good thing is, parking lots are huge opportunities for developers. I'm from DC, and the city has radically developed itself over the past few decades in turning what used to be entire blocks of surface parking lots into massive apartment and office buildings up to ten floors tall. It's very easy to develop a parking lot; that land tends to be relatively cheap to acquire and demolition costs basically nothing.
https://www.google.com/maps/place/Forsyth+St+%26+Trinity+Ave...
Atlanta's pretty run down in parts, but downtown really isn't that bad.
I have a pet theory that states that land values are higher on the coast mainly because sprawl doesn't have as many places to go, so it concentrates in smaller areas. Atlanta was historically a rail hub, exacerbating sprawl because transport could always move people farther outward. Picture pouring money onto a map, and walls are anywhere it's infeasible to develop the land. The money will spread out where there's no natural barriers. It will pile up if there's a lot of them. Money piling up means increased property values.
That's not correct. Battery Park City was developed in the mid 1970's.[1] And that development is particularly significant here as it includes a car-free green space in the form of Hudson River Park which begins there. Hudson River Park itself also being significant in that it won out over the proposed Westway Project which would have placed an interstate highway there instead.[2] The extension of Hudson River Park all the way up to the 72nd St Boat Basin is also a very recent and significant development.