Even if birds have evolved behaviors that allow the predation of noxious insects, those insect species may previously have gained benefits from their noxious venom. The arms race of prey v. predator is dynamic, not static, and evolved bird hehaviors become new selective pressures on the insects.
Everything changes. For example, South American macaws ingest clays that buffer toxic fruit: nonetheless, the toxicity of the fruit has benefits in regard to other potential consuming species, and no doubt once deterred the macaws.
Further, the existence of bird species that have evolved behaviors that allow the predation of noxious insects does not mean that the noxious venom is useless: many bird species may be deterred, even if all are not.
In addition, there are species other than birds that may be deterred.
Finally, many noxious insects are social. The complete elimination of predation on individuals may not be the point: the recognition of an erupting hive of noxious insects may evoke avoidance behaviors by predators, even if the appearance of a single such insect does not. Thus, the predation of noxious individuals may be opportunistic and of relatively trival importance to the hive.
It is tempting to oversimplify that which we wish to refute, and I believe this is what you have done.