Amyloid-associated neurodegenerative diseases, including Alzheimer’s disease (AD), are characterized by the in-brain accumulation of β-sheet structured protein aggregates called amyloids. However, neither a disease model nor therapy is established. We review past data and present new, preliminary data and opinions to help solve this problem. The following is the data-derived model/hypothesis. (1) Amyloid-forming proteins have innate immunity functions implemented by conversion to another sheet conformation, α-sheet. (2) In health, α-sheet structured, amyloid-forming proteins inactivate microbes by co-assembly with microbe α-sheets. Amyloid-forming proteins then undergo α-to-β-sheet conversion. (3) In disease, α-sheet-structured, amyloid-forming proteins over-accumulate and are neuron-toxic. This hypothesis includes formation by virus capsid subunits of α-sheets. In support, we find that 5–10 mM methylene blue (MB) at 54 °C has a hyper-expanding, thinning effect on the phage T4 capsid, as seen by negative stain- and cryo-electron microscopy after initial detection by native gel electrophoresis (AGE). Given the reported mild anti-AD effect of MB, we propose the following corollary hypothesis. (1) Anti-AD MB activity is, at least in part, caused by MB-binding to amyloid α-sheet and (2) MB induces the transition to α-sheet of T4 capsid subunits. We propose using AGE of drug incubated T4 to test for improved anti-AD activity.