* Hossenfelder's own toy model, https://arxiv.org/abs/2010.01327v5
* Gerard 't Hooft's cellular automata model: https://webspace.science.uu.nl/~hooft101/gthpub/FFP11_2010.p...
* https://www.journals.uchicago.edu/doi/10.1086/714819
Models that violate Bell's theorem are in fact simple to construct. Everyone agrees that superdeterminism can evade Bell's theorem, the key is making this evasion as plausible or more plausible than accepting many worlds or the absence of counterfactual definiteness.
Given so little effort has been expended in this direction because of incorrect assumptions of superdeterminism, it's not surprising that these models are still rudimentary proofs of concept, but the notion that Bells' theorem is some insurmountable obstacle is just incorrect.