Which is one of the main points of the article: the limiting factor is not the type of propulsion, but how and where to put the tracks.
Tunneling is expensive. Decreasing tunnel diameter just turns one knob of that system. Emergency access, switching and mergeing systems, and access to the tunnel itself use space and cost money. And if the idea is to just add tunnels to increase capacity, then that is probably more expense than building a proper train system with higher passenger density in the first place.
Trains that hang from stilts still need the stilts to sit on the ground somewhere.
Trains that sit atop of roads are limitied in their speed, by the traffic these roads where originally designed for. The curves might not be suitable to go around at high speeds.
You can't also just increase speed - acceleration and deceleration forces as well as the forces acting on a whatever kind of pod on whatever kind of track are physical limitations that affect passenger safety and comfort.
All these ideas improve on some small aspect of the system as a whole, but they do not solve the fundamental difficulties inherent in designing public transportation - they do the job, but not without the same real world problems that we have with traditional rail.
Autonomous operation for increased frequency might make some sense, but then that's something that can probably be retro-fitted to existing systems, and has little to do with the propulsion system in use.
... and I know of at least one urban public transport system where the limiting factor for increased frequency is ( apart from the issue that six train lines share one bottleneck track ) the time the doors take to close safely... not something I would have had on my radar as being a problem.
Making train tunnels cheaper by making them smaller is not a new idea. it is a very old idea. Have a look at the deep level tube lines of the London underground. In some ways, we regret having trains that small since it reduces capacity and makes it nigh impossible to air condition.
It's possible to make normal trains go over existing roads, elevated rail is also a very old thing, a lot of it did get taken down though.
Also if you want small trains, just look at the UK. Normal trains here are much narrower than normal US trains, they are about as wide as modern trams! And the London underground has trains that are tiny and shaped to make the most out of circular tunnels.
You don't need to rethink the entire system for the points you've raised there...
First of all: trains are great, there are many places where they perform admirably. Many cities have a great subway network, countries like France, Japan, Taiwan and now China have great high speed reail networks. So if there are transportation problems to solve, it should be a no-brainer to consider traditional train systems first.
However, there are shortcomings wiht traditional train systems, these new systems try to address. One big limitation for trains is, where to put the tracks. Inside cities, going underground usually is the only solution. With traditional rail, this is extremely expensive and also a slow process. In Munich, not log ago the digging for a new underground train line started - it is expected to take a whole decade and going to be quite expensive (which has delayed the project by another decade, as people were discussing costs). Outside of cities, rail is possible, but still faces problems to get the necessary ground. Noise is a further problems with trains. Finally, most traditional train systems are not operated autonomusly, so costs are too high for high frequencies and often enough there are just not enough drivers available.
These shortcomings create a lot of creativity to address them. The Boring companie tries to lower tunnel costs by making them less in diameter. Other systems try carts which run on or hang from stilts, reducing the space needed or allowing the train installations to go on top of existing roads.
Maglevs are still a huge topic. On the one side, they do allow higher speeds, as rails get mechanically challenging above 300kph. The only challenge for maglev is air resistance, there is no risk of derailing. Also, at any speed, maglevs are less noisy than rails, they can accelerate faster than rail and most importanly, go up much higher inclines. Maglev is therefore not only interesting for long-distance transportation, but also inside city regions, especially due to the lower noise and also because of autonomous operations. A good example is this maglev system: https://transportsystemboegl.com/ It is designed for city regions, fully autonomous and not just some technology concept, but ready-to-buy. It has been developed by a major construction company. That means, they won't just sell you some fancy technology, they are going to build the whole infrastructure for you ready to use. This is one of the two companies responsible for the quick building process at the Tesla Giga Berlin factory. They are basically factory producing the parts for industrial buildings and just assembling them on-site. (Ironically Volkswagen is also building a new factory for electric cars in Bremen, Germany at the moment and it looks exactly like what Tesla is building, as they ordered the factory from the same company)