It still eludes you.
That post was not about the power of reflection (how much it is visible), but about which objects gets reflected. The LFS reflection is optimized by drawing only "upper half of globe" from the car's bottom, so there are missing curbs and other ground objects in the reflected image (as they are in the bottom half of globe). The author didn't ask for LFS to make cars more shiny, but to reflect the objects on ground too. I.e. ignore the reflection "power" on those screenshots, that comment was about what is being reflected, not how-much.