As far as I know, predators who attack the fly will aim for the wings instead of the body of the fly, allowing the fly to escape while only suffering wing damage. A similar phenomenon is seen with butterflies and their colorful wing patterns, which also often have very prominent eye-spots. Predators (such as mantises) will mistake the wings for the body and attempt to grab at the wings instead because they stand out.
See for example:
Prudic KL, Stoehr AM, Wasik BR, Monteiro A. Eyespots deflect predator attack increasing fitness and promoting the evolution of phenotypic plasticity. Proc Biol Sci. 2015 Jan 7;282(1798):20141531. DOI:10.1098/rspb.2014.1531
Mantid attack behaviours on dry season (DS) and wet season (WS) forms of Bicyclus anynana , and survival outcome for arena experiments. ( a ) The ventral surface of the two seasonal forms of B. anynana . Note the differences in the ventral hindwing eyespot size. ( b ) Latency for the invertebrate predator, Tenodera sinensis , to orient on each form of B. anynana . Means±95% CI presented. The DS form took longer for mantids to detect. ( c ) Percentage of butterfly escape once attacked by a praying mantid. The DS form was much less likely to escape once an attack was initiated. ( d ) Percentage of mantid first strike on various body parts of B. anynana . The WS form was attacked more frequently on the hindwings than the DS form. ( e ) Percentage of damage observed per hindwing eyespot in the WS form only. Eyespots Cu1, Cu2 and Pc were the most damaged.
It really takes just a tiny increased odds of survival between variants for certain traits to rise to high frequency in the population. Notice that it’s the same species of butterfly that goes through different stages in wing pigmentation. It is a very plastic trait, and they can alter the pigment expression across seasons.
And how in the world such a thing could have been produced by the RV + NS mechanism?
In all likelihood some pigment was already expressed in the wings to begin with, making them stand out more than the body did(the body color of this fly seems quite pale in comparison). Natural selection then shaped the pattern, with every increment having some small survival advantage.
Flies have many different pigment patterns in their wings:
If you can understand that having pigment expressed in the wings is a trait that exists in a population, and can vary between individuals(there is variation in the trait), and it is heritable. Then if you can accept that differences in this trait can have a survival advantage, the deal is sealed. That’s basically all that is needed. Take a look at wings with patterns like the O, P, and Q types above, which seems somewhat similar to the pattern from the fly with smaller flies on it’s wings.