Can we say that the significance of structural myths is not information ("the truth") that they presumably provide, but that they instruct us how to act or behave? Behavioral output could be very adequate (even on a massive scale) even if the myth is completely bonkers.

Myths also provide experience (emotions, visions, feelings), which is a big advantage over cold "rational" communication. That makes it easy to remember and appreciate myths, and probably a person himself appreciates it more if many parts of the brain are embraced with a myth.

Myths can certainly change, and they compete somewhat as Dawkinsian memes, though the competition rules are not simple at all - you have to take into account childhood imprinting, culture, human egos, fashion drifts... The neo-liberal (but "conservative") understanding of society and economics is actually a very young structural myth. Its vigorous aadvancement suggests that... the "sure" economic imperatives of self-interest and financial profit are viruses among the fellow myths. That kind of memes were there before, gripped whole societies swiftly, but somehow they must had subsided so that the humanity could have more "naive" mind ecologies a century or few ago.

Dawkins himself gives religion as a foremost example of a viral meme. I disagree - viruses would not be so stable through centuries. I suspect greed and violence as more vivid examples of viral memes - they are certainly contagious, able to spread fast, and have disastrous potential. It is different to look otherwise from our modern mythology of high financial, political and military interests, but are people necessarily so universally obsessed with power and wealth at all times? If greed and social dangers are growing more and more wild, what did we have before? Don't we see most people being remarkably submissive towards their financial fate, limited choices, toxic status quo?