In this paper, we investigate the differential cross-sections of the inclusive jet and dijet productions of the ZEUS collaboration data at the center-of-mass energies of ∼300 GeV and 319 GeV using the k t and (z, k t )-factorizations with the different unintegrated and double-unintegrated parton-distribution functions (UPDFs and DUPDFs), respectively. The KaTie event generator is used to calculate the differential cross-sections of the UPDFs, while for the input DUPDFs, the calculations are directly performed by evaluating the corresponding matrix elements. We check the effects of choosing the different implementations of angular or strong ordering constraints using the UPDFs and the corresponding DUPDFs of the Kimber–Martin–Ryskin (KMR), the leading-order (LO), and next-to-leading-order (NLO) Martin–Ryskin–Watt (MRW) approaches. The impacts of choosing the virtualities or in the differential cross-section predictions for the ZEUS experimental data are also investigated. It is observed that, as one would expect, the application of (z, k t )-factorization is better than the k t -factorization framework for predictions of high-virtuality Q 2 with respect to the ZEUS collaboration data, and also that the results of the KMR and LO-MRW UPDFs and DUPDFs are reasonably close to each other and, in general, can describe the data. It is also observed that only in the case of k t -factorization does the inclusion of the Born level make our results overshoot the inclusive jet experimental data.