Being continuously self pollinated means the fusion of same/similar genes from time to time, then why don't they show such a decrease in productivity?
Yes, self pollination does lead to inbreeding depression:
However, we found substantial inbreeding depression, with seeds per fruit, progeny survival and progeny growth being lower after self- than after cross-pollination. Progeny arising from self-pollination also had a higher frequency of certain traits ... which were associated with lower rates of survival. ... high inbreeding depression has also been reported in other self-compatible trees ...
Inbreeding depression is one of the leading factors preventing the evolution of self-fertilization in plants.
The strategy for breeding F1 hybrid squash is to develop parental lines through self-pollination. However, it increases plant mean homozygosis, which is not the natural genetic state of a cross-pollinated species, and can cause "inbreeding depression".
So why don't all crops that self-pollinate show decreases in productivity? In part, they do, but the reduction in productivity caused by inbreeding is outweighed by the increases in productivity caused by inbreeding (i.e. selection for productive traits via inbreeding leads to both positive and negative results, and the total outcome is the sum of the positive and negative effects).
And partly because the deleterious effects of inbreeding can be partially overcome through "purging". The negative genes that lead to inbreeding depression when homozygous can be removed from a population as the affected individuals die (or fail to breed) and take their negative genes with them. Purging is most common when a population has a moderate degree of inbreeding over a longish time, which is what you'd expect with a crop that typically self-pollinates. It's an effect in many but not all common crops:
Several plant species, E. g alfalfa (M. sativa) carrot (D. carota) , hayfield, tarweed etc show very high inbreeding depression. ... Many crops species, such as maize, jowar, bajara etc. shows moderate inbreeding depression. ... Several crop plants, E. g onion (A. cepa), many cucurbits, rye (S. cereale), sunflower (Hannus), hemp etc show only a small degree of inbreeding depression. Only a small proportion of the plants show lethal or subvital characteristics. The loss in vigour and fertility is small; rarely a line cannot be maintained due to poor fertility. The reduction in yield due to inbreeding is small or absent. Some of the inbreds lines may yields as much as the open pollinated varieties from which they were developed.
Interesting question, but the answer might not be very simple. I'm not a big expert, but will try to give some clues.
So, first of all in many cases there is inbreeding depression. For example, some crops that are highly inbred (like tomato for instance) have a big problem of genetic diversity, which is very low among cultivated variants. This is sometimes solved by introducing introgressions from wild relative species. Otherwise you end up with crops that suffer from low biotic and abiotic stress tolerance.
Also note that in some cases, the seeds that are sown by farmers are not inbred lines, but rather the F1 result of a cross between two inbred lines. This is common for instance in maize, where all commercial cultivars sold by seed companies are hybrids. This leads to a phenomenon called Heterosis or Hybrid vigor, in which the progeny of the cross (the hybrids) have improved phenotypes over the parental lines.