Is UAC an overestimated protection measure in Windows? Does Linux do it better?

Question

Windows is an API-based operating system, whereas Linux is file-based. Hence in my opinion, any operating system API security measures alone aren't effective in Windows. One of such examples would be UAC in my humble opinion. On the other hand, Linux incorporates SELinux and sudo instead. This goes to low-level layers of the operating system, which are harder to manipulate in my humble opinion. The Windows low-level approach (on older Windows Server systems at least) of pressing Alt+Ctrl+Del to login was a good approach in my opinion, but doesn't protect at all in this scenario in my opinion.

In my opinion, UAC only protects against a malfunctioning (digitally OR physically) mouse, which double clicks instead of single clicking and primitive forking malware/fork bombs.

• Is this the reason Linux has no real UAC equivalent, but rather other and actually effective solutions?

• Does any compliance measure ever recommend to turn off UAC due to it being nearly useless?

• Is it technically smart to disable UAC completely if your threat model is not protecting against a malfunctioning mouse which clicks twice with a single click or primitive malware which forks itself and does not bypass UAC?

• Does the first to third paragraph contain any misconceptions/false information?

• Does UAC protect against BadUSB-based keystroke injection? No, because you can just click yes and no password is required, right? Or is there some randomization in play here, which a Ducky Script would fail at?

• Are UAC bypasses common on recent Windows systems? Supposedly there's a backdoor on default settings and at least(!) 77 ways to do it: https://github.com/hfiref0x/UACME

• Does Linux solve these problems better?

• According to https://security.stackexchange.com/a/81770 the X window system is vulnerable to keyloggers, whereas Windows with UAC isn't, is this true? If so, does Wayland do better?