Why do you say that the technique is potentially "insecure" ? This would only be used to verify signatures on the blockchain, how can this be "insecure" ?
The signature verification can be tested thoroughly against a reference implementation (OpenSSL and sipa), so the likelihood to incorrectly verify signatures (either declaring them valid, when they are not, or vice-versa) is very low.
The method will definitely pass valid signatures.
I was concerned that there might be a way to sign invalid signatures somehow.
From
Google, the authors have presented a system which uses "Randomisers". That suggests that their original scheme had some potential (or actual) weakness.
Verify
A = B
C = D
is weaker that just verifying
A + C = B + D