Here's one more possibility to reduce the schedules without reducing the learning: take into account our practice of Japanese texts besides the quiz questions themselves.
A first way to do this could be to have tick buttons next to example sentences (in the dictionary, in vocabulary quiz after answering, in sentence questions, in grammar questions) to signify that you read the sentence, fully understood its meaning, and understood each word (possibly, after clicking on unknown (underlined) words to either see their meaning or mark them as known). Then the known words of that sentence that are scheduled don't go up in mastery, but the next ①term→meaning and ②kanji→kana questions are somewhat pushed back, to take into account the fact that you just refreshed them in your brain. It would need some experimenting, but say you start by pushing back those questions by 30%: You clicked to mean that you fully understood a question containing a word which was last quizzed (for ① and ② questions) 10 days ago and was planned to be asked again 10 days from now. Then instead of that planned 20 days interval for this word, it's now 26, and the next question is 16 days from now.
After one or 2 months of such an experimental feature with some brave volunteers, you'll start to have plenty of data, and you can compare the rate of correct answers for the words that were thus pushed, in the schedule, from the rate of correct answers for the other words in the same schedule, for the same user. If the "normal" correct answer rate is 85%, and the rate for pushed-back words is 90% or more, you'll know it works even better than expected and you can think about increasing this 30%, maybe to 40%. If the rate for pushed-back words is around 85%, it'll be a complete success, if it's 75%-80%, then prolly increasing the intervals by 20% (instead of 30%) is enough. And if the result is lower than 75% then prolly the experiment failed... But you can't tell until you try.
Another way to push back the questions could be (still with the same example) to change the date of the last ① and ② questions for that word from 10 days ago to today, meaning the next question is now 20 days from now, instead of 10 (or 16 with the other calculation). After all, if you just read that word, and understood it, it means you just successfully answered the ① and ② questions for this word.
If it works, basically the ① and ② questions for common words will almost completely stop to appear in quiz questions because they'll keep being pushed back again and again just from seeing them in context. And the kind-of-common words will be pushed back frequently enough. Overall, this could significantly reduce the size of the schedules and keep them manageable, and give more time to practice actual Japanese in context.
And if it works, then it could be extended to NHK Easy and Hukumusume? (1 tick button per sentence) Both are published with furigana, so you can probably let the parser parse them automatically, using the hints of the furigana.
Android apps offering the NHK Easy news with translations of each word are a dime a dozen (and most of them have 10,000+ downloads, 3 or 4 have 100,000+), but none of them makes a really intelligent, practical use of the texts to integrate the vocabulary into a systematic learning strategy / SRS. One of them kind of does, but other than this feature, it's no match at all for Renshuu.
So not only this could greatly help your current users spend less time on quizzes for an equivalent learning value, but also you can expect quite a crowd once you reach the Google store (any soon?) and "NHK Easy News" is in the description. And if you decide to let only the "pro" accounts tick the NHK/Hukumusume sentences, well, that would be quite an incentive, because it would be one more unique and very interesting feature of Renshuu pro…