To use your myostatin mutation example, the increase in muscle mass does not lead to significantly faster animals as the supporting structures are not there to utilise this increased muscle. Human intelligence is an emergent trait like speed determined by the co-ordination of many sub-systems.
These sort of emergent traits are almost never positively controlled by single large positive alleles. The one major exception are systems that are under different selection in males and females (e.g. height, plume color in birds, etc). In these examples two different systems have emerged controlled by the sex of the individual, but you don’t tend to find single genes that contribute massively in a positive way towards variation within each sex.
As an aside I remember reading a paper from long ago that suggested that high intelligence was the result of the relatively absence of mutant alleles at the various intelligence loci. When you look at the effect of null mutations at these loci the effect on intelligence is very low (less than a point for most). We all carry a large number of mutant alleles so the suggestion is that those with a high intelligence just happen by chance to have a lower frequency of negative mutant alleles. The interesting thing about this hypothesis is it would suggest that high intelligence is the default and low to normal intelligence is the result of mutational load. This is defiantly something that we can explore in the future as we get whole genome data from large numbers of people.