USB-C isn't a utopia in terms of fixing the USB complexity.
All it truly solved was the connector problem which by itself is HUGE in retrospect and not something I appreciated.
If you thought the transition from ATA -> SATA or PCI, PCI + AGP, PCI-E path was bad, imagine trying to move an industry like USB that has USB1.0, USB2.0, USB3.0, USB3.1 and deals with peripheral components that change at an even slower cadence than internal ones (computers are something like 3-5 years, peripherals are like 3-10). By comparison mobile has had a life span of ~1-3 years (slowing down I think as there's plenty of 3-5 year users now & their numbers grow every year). The 1-3 cycle only happens for as long as you have a growing user-base & the churn is required for rapid change to be possible (the market economics support it).
When I think about the market dynamics, USB-C is a critical backbone you had to land first (which I didn't see when it was first being pushed on us in the consumer electronics industry). USB4 is going to clean some of it up. Without it we were never going to get away from the mess we were in. That being said, USB-C is still a big too big. That's why Apple hasn't moved. The question will be if the next push will be for contactless data transfer or a smaller connector. It'll depend on whether mobile needs the space (& again - even mobile is losing its power to power such a transition) & whether major consumer electronics will identify new small form-factor markets that require a smaller connector. If the Apple Watch/AirPods is any model for how these will work, then no. So we'll get USB5 & USB6, but it'll be "boring" transitions like it is with Ethernet RJ45 cables (Cat 4, Cat 5, Cat 6, Cat 6e, Cat7) etc. Some of the customer complexity will be trickier to navigate if the labelling piece isn't addressed for cables & ports (otherwise you get an alphabet soup of standard names supported by a given cable/port & what the peripheral requires). These are hopefully more manageable problems though (although more tricky than Ethernet). One could imagine where you have different classes to describe power the cable & port can run, a different class of the functionality supported (power only, peripherals, peripherals + displays). Peripheral speeds here would be > than ethernet so we're also potentially going to see USB-C consume the space of ethernet ports. This happened in laptops but it'll be trickier in PC form factors. Everything you'd plug into is RJ45 anyway & ethernet is significantly cheaper on a $/ft basis since it's cheaper. Maybe it'll intercept when consumer ethernet starts to be 10gigabit/s or 100gigabit/s (if USB4/5/6/ supports the speed & lengths needed & the cost is comparable, they'd probably just layer it over that). I remember working with 1gigabit and 10gigabit/s in 2009 at QCOM when the Cat6 standard for Gigabit had just come out & gigabit has only became prevalent in the past few years. You naturally want to consolidate these things as a business to simplify your costs if your sell-through is into a market with very long cycles. Any market that has more iteration. Wireless comms will be that a for a bit until BT & WiFi eat everything they haven't already (& potentially WiFi will eat BT eventually). Proprietary cheap connectors/wireless in the ultra low-cost niches will be the longest holdouts but even then they'll struggle as the customer demand will probably prefer the standard name it sees everywhere else.