Antimicrobial peptides (AMPs) display advantages over traditional antibiotics due to their broad spectrum of activity against various pathogens, and may even overcome bacterial drug resistance. However, despite their potential therapeutic benefits, widespread application of AMPs is limited by their instability, sensitivity to high salt concentrations, toxicity, and immunogenicity. Lipidation is a promising strategy in overcoming these drawbacks and potential problems for drug candidates. While N-terminal lipidation is a well-studied form of acylation of biologically active peptides, fatty acylation of the lysine side chain has still been poorly explored. In this study, we examined systematic introduction of octanoic (C8) or decanoic (C10) acid into the sequences of three antimicrobial α-helical peptides, namely LL-I, LK6, and ATRA-1, by acylation of subsequent lysine residues, resulting in 17 lipopeptides. Fatty acid lengths optimal for antimicrobial activity were selected based on a previous study on the N-terminal lipidated counterparts of these peptides. Shuffling the position of the fatty acid tails in the sequences of the peptides preserved high activity against Gram-positive bacteria, increased activity against Gram-negative strains and reduced cytotoxicity, compared to the N-terminal acylated counterparts. In the case of the LL-I and LK6 conjugates, the interactions with artificial negatively charged membranes induced formation of an α-helical structure but without a direct correlation between helicity and amphipathicity. Unexpectedly, the ATRA-1 derivatives showed only a small tendency, if any, to adopt a helical structure upon binding to POPG vesicles, which may indicate a non-helical active conformation. A more detailed study of the selected analogues, namely LL-I-4C8, LK6-7C8, and ATRA-1-11C10, provided evidence of a tendency to self-assemble into clumped and/or isolated fibrils, micelles or clusters of micelles, and proved that the lipid bilayer is the main target of action of the tested lipopeptides. In summary, the results of the present study highlight that alternative conjugation sites for lipid modification of AMPs, rather than the commonly applied N-terminal conjugation site, may improve the selectivity of action and be feasible in testing for the development of new lipid-peptide conjugates.