if you are intent on playing the K 467 first, you can also listen to murray perahia's. they are very good. someone else mentioned cassedesus - who was also recommended. the problem is, finding these cadenzas or getting them transcribed off cd. i don't know who does this. i was looking for someone to do that very thing for my 'piano concerto' class. badura-skoda's was cut and pasted by ashkenazy, i think. it sounds ok. but i like murray perahia's the best. short and sweet. mozart's were fairly short, usually. beethoven gets wild and modulates farther than he should - and lengthens the cadenzas with more of a 'development' than mozart did.
barenreiter urtext has all the mozart cadenzas extant (i think) - but you can also imagine from reading the forward that some have probably gone missing. the idea that he always extemporaneously played cadenzas isn't totally true. he often had folios of his candenzas and kept them closely guarded. wouldn't doubt that some of these missing ones were lost - but - i believe they existed at one time. he was very particular about writing down the basic ideas of his cadenzas, imo. even when he himself was playing. and he distinguished in recital programs the difference when he was 'extemporizing' and performing a concerto. even if he was using the basic chord progressions - he had at least a sketch.
i wrote a paper on mozart cadenzas and i can't find it now. but, basically mozart was not stylistically excessive. he followed the general plans of the day (dg turk's rules for writing cadenzas: klavierschule, leipzig 1789) and uses his own harmonic cycles. they are short, have only bits of thematic material, have a good deal of sequence and figuration, do not use modulation (terribly), and use 'harmonic cycles.' the subdominant regions are always elaborated after the 'interruption' (long held note- tremolos) and not before. mozart gives a fine example of working in a composer's own themes into the cadenza with his writing of several cadenzas for js schroeter (a contemporary) although he also uses some of the composers own stylistic manners - so you cannot tell if he would use a similar cadenza in his own work. but, in terms of a 'cross-reference' the K 626 AII - kochel writes 'piano cadenzas of mozart - not related to his own concerti.'
and, of course, you have all the extant cadenzas written for his own concerti. that would be the safest way to go. to play some of the ones that DO have cadenzas and get used to how they sound and what is 'appropriate.' mozart's 'a musical joke' jested about how music can be 'second rate, superficial, stubbornly and enduringly devoid of any significant idea.' he violates elementary laws of composition by creating consecutive fifths and octaves, doubling parts without account for texture, creating overly intrusive accompaniment in sections, and music that goes nowhere and cannot come to the cadence appropriately. when the cadenza arrives for the violin, it is overtly an example of how a cadenza should not be written.
arthur hutchings says 'beethoven's cadenzas are too interesting.' and, perhaps it indulges in rather too many moods and illuminations, and thus has a retarding effect where an enormous pile-up of tension makes a fierce release indispensable (quote from badura-skoda). many people wrote for the K 466. i don't know who wrote for the K 467. for the K 466 we have brahms and clara schumann's, too. john cramer wrote cadenzas for K 450, 467, 482, 466, 491). emanuel forster wrote pairs of cadenzas to three mozart piano concertos and sounded enough like mozart to have been requested by andre on a certain occasion. heinrich henkel wrote one cadenza for K 503. karl hoffman wrote cadenzas for K 467, 482, 488, 491, 503, 595 - but they were not favorably revewed in 1801 in vienna as they were thought to be overly massive and had overcharged figuration. johann hummel composed for K 491, 595, 415, 413, 537, 459, and 414. arthur hutchings thought they were also of excessive lengths. especially K 491's. although some think the ones he wrote for the K 491 are the finest ever written. ie david grayson. englebert humperdink wrote one for the K 491. felix mendelssohn wrote for K 466, K 365. audiences seemed to like them. (he may have written more?) mozart's own son franz xavier mozart K 466, K 460, K 467, and K 503. they are preserved in the bibliothek des conservatoriori giuseppe verdi. august muller wrote to 8 concertos. he was the former cantor of the church of st. thomas and had been apupil of jc bach. carl reineke composed cadenzas to most of mozart's pc's. published them under op. 87. camille saint-saens composed cadenzas to three mozart concerti but they are too foreign in style to be recommended. aloys schmitt wrote cadenzas for third movement of K 466 for two pianos. clara schumann's story is interesting. bedrich smetana composed some for K 466 and K 595. he was chosen to play mozart's d minor concerto for the 100th anniversary in prague celebrating mozart's birth.
what i like about murray perahia's is that they are 'sparkling and effervescent' rather than long and tedious.