Theorem 0.2 in the author's paper [14] asserts that a Lagrangian Klein bottle in a projective complex surface must have non-zero mod 2 homology class. A gap in the topological part of the proof of this result was pointed out by Leonid Polterovich. (It is erroneously claimed in §3.6 that a diffeomorphism of an oriented real surface acts in some natural way on the spinor bundle of the surface.)Recently, Vsevolod Shevchishin corrected both the statement and the proof of that theorem. On the one hand, he showed that the result is false as it stands by producing an example of a nullhomologous Lagrangian Klein bottle in a bi-elliptic surface. On the other hand, he proved that the conclusion holds true under an additional assumption which, in retrospect, appears to be rather natural. Shevchishin's proof of Theorem A uses the Lefschetz pencil approach proposed in [14], the combinatorial structure of mapping class groups, and the above description of symplectic manifolds with c 1 (X, ω)·[ω] > 0. The purpose of the present paper is to give an alternative proof in which the first two ingredients are replaced by somewhat more traditional four-manifold topology. It should be noted that though closer in spirit to the work of Polterovich [17] and , this argument has been found by interpreting the results obtained in [18] in other geometric terms.The contents of the paper should be clear from its section titles. For a more comprehensive discussion of Givental's Lagrangian embedding problem for the Klein bottle [6], see the introductions to [14] and [18].The author is grateful to Viatcheslav Kharlamov for several helpful remarks.