You want the change to take a long time so that humans can evolve with it. Yes plant hybridization (e.g. to make fruit sweeter) is not as fast or easy which is why it's more in tune with evolution. Some things that genetic modification can do are impossible with the ancient methods.
Even if GM produced a plant that was truly superior and healthier in every way, there's no good way to trust it. There are too many instances of "oops, what we thought was good for you is actually bad for you" or "oops, we thought we could improve nature but we made it worse" to trust GMO, at least for the next fifty years or so. The plant that seems truly superior and healthier might end up killing enough insects or otherwise affecting them such to be worse all things considered.
Evolution takes much longer time than selection. And human evolution does not function like it did since humans have medicine and society, which means if you can't eat grains, for example, you don't die and take your genes out of gene pool - you just take some medicine and eat something else and live happily and procreate. So evolution has nothing to contribute with regard to organisms modified by human selection - it's both too fast and evolution doesn't have a chance to do its cruel business.
Im modern mind "evolution" has became synonymous to "natural" in the quality of "it has no nasty science in it so it has to be good". In fact, neither "natural" nor "evolution" has nothing good in it - nature is impersonal, immoral and unimaginably cruel, viewed from our point of view. And if the science can do something to correct it - it is great.
For example, sweeter apples we enjoy now appeared in the last couple of centuries, and the US varieties in the last century alone, before that apples were mainly used to make cider or applejack. A couple of hundreds of years is nothing for the evolution.
Yes from our point of view, where human death is always bad and future consequences are largely ignored. From a macro point of view nature achieves what is perhaps the best possible outcome. Man has a spotty track record of improving or correcting it. We made some improvements that still seem better over the last century, like vaccines. And we've made many big mistakes, like DDT. What you support is a matter of what seems trustworthy.
Search for: "Pests are adapting to genetically modified crops in unexpected ways, researchers have discovered. The findings underscore the importance of closely monitoring and countering pest resistance to biotech crops." That sounds like another "oops, we did it again" in the making to me.
Science is not perfect. It's the best we have. Yes, pests are adapting and we don't have silver bullet. The solution is, however, not to say "well, science didn't work out, let's try to go back to the caves" - the solution is to keep working.
>>> What you support is a matter of what seems trustworthy.
This is exactly the problem - the support becomes political and relies not on what is actually better but on whose lobbyists are better and whose PR campaign is more successful. This venerable tradition started a long time ago, with Edison's shenanigans[1] to give him edge over the competitors. The story of DDT is another example - while the scientific basis for the book that destroyed the DDT - Silent Spring - remain shaky[2] and the ban ended up costing millions of lives of malaria victims, it is still considered an example of how science is going to doom us all. Another example is the continuing PR battle between sugar lobby and corn lobby. Initially, corn lobby was winning - that's why HFCS is called "high fructose" - because once fructose was good and glucose was bad, so the marketing term emphasized the good part, even though some varieties of HFCS have less fructose than table sugar, and commonly used one has only very slightly more [3]. Due to this and as the effect of the insanely high sugar tariffs, US industry switched from sugar to HFCS. Now the tide has turned and HFCS is becoming the bane of our age. And now the corn lobby wants to go back to "corn sugar", but so far they weren't allowed to. Corn guys though did another number in "biofuels", so even if they lose the HFCS thing, they have other victories in their belt.
This is all to say "what seems trustworthy" is a very poor guide, since most information you'd see in popular press is probably worthless. And decisions made by politicians are probably based on random factors having nothing to do with actual merits of proposed solutions.
Even if GM produced a plant that was truly superior and healthier in every way, there's no good way to trust it. There are too many instances of "oops, what we thought was good for you is actually bad for you" or "oops, we thought we could improve nature but we made it worse" to trust GMO, at least for the next fifty years or so. The plant that seems truly superior and healthier might end up killing enough insects or otherwise affecting them such to be worse all things considered.