That is a really good question. Good enough that each dam should be evaluated on a case by case basis. In the instant case of the Klamath dams, the lower 3 don't store much water. The upper dam that backs water into Oregon is a shallow, but large in acreage reservoir. I'd have to look up its storage if we want to go there. I've read that Klamath irrigation water is mainly used to grow hay. I'm not sure how much is used to grow crops for direct human consumption. I have no idea if it's true, but I read that the hay is exported to Japan (and maybe elsewhere) to feed race horses. Nonetheless, growing hay should rank lower than human food crops. (Raising Wagu beef is a long ways from saving humanity.)
We should note here, that in CA, a state of 40 million people, 80% of all water used, is used by agriculture. Municipal and industrial (M&I) use accounts for 20%.
Comparatively, removing the Klamath dams won't reduce the energy supply noticably or the water supply for human use at all. Water for irrigation will be reduced, but I don't know what the impact will be. A lot of the farmers use flood irrigation, the cheapest and least efficient method. Upgrades to irrigation methods would ensure that the loss of crop production would be less than the proportionate loss of water available to irrigation. And as Mr. Smith pointed out above, the Klamath dams are responsible for detrimental effects on water quality.
Moving south, we have the perimeter "ring dams" of CA's Central Valley on the Sacramento and San Joaquin Rivers and their tributaries. Many of these dams have reservoirs that store huge amounts of water. I feel confident that, fish or no fish, none of these major storage dams are going to be removed. Some of them produce significant amounts of energy, and they provide a huge amount of water for agriculture, M&I use, and numerous drinking water supplies beyond the M part of M&I. That water supply is critical to the survival of CA, making it worthwhile to spend ridiculous amounts of money on fish passage and features like multi-level water intakes that improve water quality where needed. Similarly, farmers will be paid to install more efficient irrigation systems so they can sell surplus water they create to M&I users. I can envision a future where CA farms use Israeli style irrigation - Israel grows food in a desert. I think it's safe to say that they know a thing or two about efficient irrigation.
Long answer to your question, but yes, it matters how much water a dam stores, and it probably matters even more where that water reservoir is located. And it will matter how efficiently that water can be used for energy production, irrigation, and M&I use. It shouldn't be a one size fits all, but rather a case by case evaluation of costs and benefits and the values we place on those benefits. Will society value Japanese race horses over native spring Chinook?