The SFO case is interesting, because it's not just "foreign crews don't like visual approaches". It's because the US is commonly using visual approaches kind of as a loophole to reduce separation. All major carriers require that crews use the ILS (or RNAV) as a backup to a visual approach if it's available. But by clearing for a visual approach the controller is no longer required to keep the planes on approach separated.

I know a few European airlines that encourage their crews not to accept visual approaches in the US and request an ILS approach instead, because it's safer.

Europe has solved the same "we need more planes per hour" problem by allowing a reduced 2.5nm separation when both planes are established in the localizer and are radar monitored by the controller.

As a pilot I prefer the European way, because it's more structured and has a few more safeguards. The US method puts the problem with the pilot, who has far worse means to judge separation than the controller