OTL the kingdom was only a few fortified coastal settlements and cities but it also only lasted twenty years. One could easily compare this to Venice control over Dalmatia and Greece which likewise waxed and waned.
This is true, and it's not that bad of a comparison - the Dalmatian cities were also "protectorates" with little Venetian administrative interference for a fair period of time. But the differences are also considerable. The Dalmatian city-states were, in a sense, part of the same realm as Venice; they were all Byzantine satellites, and indeed Venice's dominion there sprung from an imperial grant. They shared a religion as well as a culture and mode of living.
The issue here, to me, is that it's difficult to see real cultural interchange happening on the ground. There's no reason to think that there would be serious immigration from Europe, and while you might see mercantile enclaves set up in the client cities of the coast that's a more limited form of population mingling. (Even Alexandria had Italian merchant quarters, after all.) As for the interior, the best that the Sicilians can probably do is to play intermediary, supporting one tribe/confederation against another. That's a role they can fill pretty well, especially if they control the entrepôts of the coast and thus the flow of much of the region's wealth, but it's not a role which lends itself to the actual annexation, let alone settlement, of the inland regions which one would need possession of to reverse the "pastoralization" of the region which accelerated after the Bedouin migration.
My opinion is that if the Almohad conquests are somehow averted and "European Sicily" is preserved from other threats such that the Kingdom of Africa endures, it is most likely to take the form of a string of vassal city-states, perhaps ruled by
ammirati (emirs) or
gaiti (from
gaitus, the Norman-Latinization of
qaid), who over time might adopt Norman titles or perhaps even Christianize, but whose populations are going to be rather mixed, consisting of settled Berbers (probably Muslim), Arabized city-dwellers of both religions, pre-conquest Christian remnants, African Jews, and Italians (principally concerned with trade, I'd imagine). I suspect the establishment of this protectorate might paradoxically
accelerate the de-Christianization of northern Africa, as Christians would conceivably relocate to the coastal cities where their religion was not only not persecuted but favored. This seems to be exactly what happened later on IOTL, when Christians in the later Middle Ages converged in a handful of cities in order to exist in communities of sufficient number to maintain their religious and communal life.
Beyond the immediate environs of these cities, the Berber and Arab Bedouin tribes are likely to remain Muslim, but ally provisionally with the Normans in order to gain the benefits of trade and keep that wealth from the hands of their local competitors. That will remain the best way to "secure the hinterland;" I can't see an actual occupation of any substantial part of the Maghreb as either tenable or desirable for the Sicilian state. There's really nothing to "crusade to" once you secure the coastline anyway.