This paper introduces a new one-more computational problem for lattice-based cryptography, which we refer to as the Algebraic One-More MISIS problem, or \(\textrm{AOM}\text {-}\textrm{MISIS}\) for short. It is a modification of the AOM-MLWE problem recently introduced by Espitau et al. (CRYPTO ’24) to prove security of new two-round threshold signatures. Our first main result establishes that the hardness of \(\textrm{AOM}\text {-}\textrm{MISIS}\) is implied by the hardness of MSIS and MLWE (with suitable parameters), both of which are standard assumptions for efficient lattice-based cryptography. We prove this result via a new generalization of a technique by Tessaro and Zhu (EUROCRYPT ’23) used to prove hardness of a one-more problem for linear hash functions assuming their collision resistance, for which no clear lattice analogue was known. Since the hardness of \(\textrm{AOM}\text {-}\textrm{MISIS}\) implies the hardness of AOM-MLWE, our result resolves the main open question from the work of Espitau et al., who only provided a similar result for AOM-MLWE restricted to selective adversaries, a class which does not cover the use for threshold signatures. Furthermore, we show that our novel formulation of \(\textrm{AOM}\text {-}\textrm{MISIS}\) offers a better interface to develop tighter security bounds for state-of-the-art two-round threshold signatures. We exemplify this by providing new proofs of security, assuming the hardness of MLWE and MSIS, for two threshold signatures, the one proposed in the same work by Espitau et al., as well as a recent construction by Chairattana-Apirom et al. (ASIACRYPT 2024). For the former scheme, we also show that it satisfies the strongest security notion (TS-UF-4) in the security hierarchy of Bellare et al. (CRYPTO ’22), as a result of independent interest.